Predicting the difficult laparoscopic cholecystectomy: development and validation of a pre-operative risk score using an objective operative difficulty grading system

Abstract

Background
The prediction of a difficult cholecystectomy has traditionally been based on certain pre-operative clinical and imaging factors. Most of the previous literature reported small patient cohorts and have not used an objective measure of operative difficulty. The aim of this study was to develop a pre-operative score to predict difficult cholecystectomy, as defined by a validated intra-operative difficulty grading scale.

Method
Two cohorts from prospectively maintained databases of patients who underwent laparoscopic cholecystectomy were analysed: the CholeS Study (8755 patients) and a single surgeon series (4089 patients). Factors potentially predictive of difficulty were correlated to the Nassar intra-operative difficulty scale. A multivariable binary logistic regression analysis was then used to identify factors that were independently associated with difficult laparoscopic cholecystectomy, defined as operative difficulty grades 3 to 5. The resulting model was then converted to a risk score, and validated on both internal and external datasets.

Result
Increasing age and ASA classification, male gender, diagnosis of CBD stone or cholecystitis, thick-walled gallbladders, CBD dilation, use of pre-operative ERCP and non-elective operations were found to be significant independent predictors of difficult cases. A risk score based on these factors returned an area under the ROC curve of 0.789 (95% CI 0.773–0.806, p < 0.001) on external validation, with 11.0% versus 80.0% of patients classified as low versus high risk having difficult surgeries.

Conclusion
We have developed and validated a pre-operative scoring system that uses easily available pre-operative variables to predict difficult laparoscopic cholecystectomies. This scoring system should assist in patient selection for day case surgery, optimising pre-operative surgical planning (e.g. allocation of the procedure to a suitably trained surgeon) and counselling patients during the consent process. The score could also be used to risk adjust outcomes in future research.

Full text

Vol.:(0123456789)
1 3
Surgical Endoscopy
https://doi.org/10.1007/s00464-019-07244-5
Predicting thedicult laparoscopic cholecystectomy: development
andvalidation ofapre‑operative risk score using anobjective
operative diculty grading system
AhmadH.M.Nassar1· JamesHodson2· HweiJ.Ng1· RaviS.Vohra3· TarekKatbeh1· SamerZino1·
EwenA.Griths4,5,6 on behalf of the CholeS Study Group, West Midlands Research Collaborative
Received: 16 July 2019 / Accepted: 28 October 2019
© Springer Science+Business Media, LLC, part of Springer Nature 2019
Abstract
Background The prediction of a difficult cholecystectomy has traditionally been based on certain pre-operative clinical and
imaging factors. Most of the previous literature reported small patient cohorts and have not used an objective measure of
operative difficulty. The aim of this study was to develop a pre-operative score to predict difficult cholecystectomy, as defined
by a validated intra-operative difficulty grading scale.
Method Two cohorts from prospectively maintained databases of patients who underwent laparoscopic cholecystectomy
were analysed: the CholeS Study (8755 patients) and a single surgeon series (4089 patients). Factors potentially predictive
of difficulty were correlated to the Nassar intra-operative difficulty scale. A multivariable binary logistic regression analysis
was then used to identify factors that were independently associated with difficult laparoscopic cholecystectomy, defined
as operative difficulty grades 3 to 5. The resulting model was then converted to a risk score, and validated on both internal
and external datasets.
Result Increasing age and ASA classification, male gender, diagnosis of CBD stone or cholecystitis, thick-walled gall-
bladders, CBD dilation, use of pre-operative ERCP and non-elective operations were found to be significant independent
predictors of difficult cases. A risk score based on these factors returned an area under the ROC curve of 0.789 (95% CI
0.773–0.806, p < 0.001) on external validation, with 11.0% versus 80.0% of patients classified as low versus high risk hav-
ing difficult surgeries.
Conclusion We have developed and validated a pre-operative scoring system that uses easily available pre-operative variables
to predict difficult laparoscopic cholecystectomies. This scoring system should assist in patient selection for day case surgery,
optimising pre-operative surgical planning (e.g. allocation of the procedure to a suitably trained surgeon) and counselling
patients during the consent process. The score could also be used to risk adjust outcomes in future research.
Keywords Surgery, Laparoscopic, Cholecystectomy· Operative difficulty· Difficulty grading, difficult cholecystectomy·
Predictive score
and Other Interventional Te
chniques
Electronic supplementary material The online version of this
article (https ://doi.org/10.1007/s0046 4-019-07244 -5) contains
supplementary material, which is available to authorized users.
* Ewen A. Griffiths
ewen.griffiths@uhb.nhs.uk
1 Department ofSurgery, University Hospital Monklands,
Lanarkshire, Scotland,UK
2 Institute ofTranslational Medicine, University Hospitals
Birmingham NHS Foundation Trust, Birmingham, UK
3 Trent Oesophago-Gastric Unit, Nottingham University
Hospitals NHS Trust, Nottingham, UK
4 Department ofUpper Gastrointestinal Surgery, University
Hospitals Birmingham NHS Foundation Trust, Birmingham,
UK
5 Institute ofCancer andGenomic Sciences, College
ofMedical andDental Sciences, University ofBirmingham,
Birmingham, UK
6 Department ofUpper Gastrointestinal Surgery, Queen
Elizabeth Hospital Birmingham, Area 6, 7th Floor,
Mindelsohn Way, Edgbaston, BirminghamB152WB, UK

Surgical Endoscopy
1 3
When performed by suitably trained surgeons using a
standard technique, the majority of laparoscopic cholecys-
tectomies are relatively easy and uncomplicated. However,
certain operative findings can hinder various steps of the
operation, increasing its complexity and leading to a higher
incidence of adverse outcomes [1, 2]. Factors traditionally
associated with operative difficulty include the presentation,
patient factors, clinical examination, blood results and radio-
logical findings. The prediction of difficult cholecystectomy
would, therefore, offer advantages to the safety of the proce-
dure and to improving outcomes.
Many studies have been published proposing pre-oper-
ative scoring methods for predicting a difficult cholecys-
tectomy. However, most scoring systems lack objective
definitions of the difficulty encountered at the time of a
cholecystectomy [3–13], being correlated to different meas-
ures of intra-operative difficulty [3, 4, 6–8, 11, 14–16].
These studies either had small sample sizes or lacked exter-
nal validation. Other predictive scores calculated the risk of
conversion to open surgery during cholecystectomy [3–6,
8–13]. However, this is a variable depending on surgeon
experience and equipment availability. In addition, as a vari-
ety of techniques and strategies are now available, allowing
for the continuation of a laparoscopic approach where chal-
lenges arise, conversion to open surgery has become less
relevant today than when the procedure was in its infancy.
A difficulty grading scale, based on specific intra-opera-
tive findings was described by Nassar etal. [17]. The com-
prehensive nature and simplicity of this grading system led to
it being utilised in multiple studies of the suitability for cer-
tain techniques and of intra- and post-operative outcomes of
laparoscopic cholecystectomy. We recently found that higher
operative difficultly score was associated with worse clinical
outcomes, including increased length of stay, complications,
conversion to open surgery and 30day mortality [18].
The aim of this study was to develop and validate a pre-
operative score to predict operative difficulty, as defined by
a validated intra-operative difficulty grading scale.
Methods
Two cohorts from prospectively maintained databases of
patients that underwent laparoscopic cholecystectomy were
analysed.
CholeS cohort
The CholeS study was a multicentre, prospective population-
based cohort study of the variations in practice and out-
comes of cholecystectomy [18–20]. The protocol did not
require research registration, as anonymous, observational
data were collected. The study was registered as a ‘clinical
audit’ or ‘service evaluation’ at each participating hospital,
under the supervision of a named senior investigator (con-
sultant surgeon). This database included 8755 laparoscopic
cholecystectomies collected from 166 hospitals across the
United Kingdom during the months of March and April
2014. The data were reviewed by independent data valida-
tion and was found to be 99.2% accurate.
Reference cohort
This database included 4089 consecutive laparoscopic
cholecystectomies performed by a single surgeon
(AHMN), or by his trainees under direct supervision, in
four hospitals between February 1992 and July 2014. It
was registered as a clinical audit in each hospital and did
not require Institutional Review Board’s approval. Accord-
ing to hospital protocols, the biliary firm accepted all
patients admitted acutely and proven to have benign biliary
pathology. All patients deemed fit for surgery underwent
laparoscopic cholecystectomy, with routine intra-operative
cholangiogram (IOC) and, if indicated, common bile duct
exploration (CBDE) during their index admission. Mag-
netic Resonance Cholangio-Pancreatography(MRCP) was
not routinely used as part of the diagnostic protocol, and
pre- operative ERCP was not relied upon for pre-operative
endoscopic clearance of bile duct stones, except in those
unfit for general anaesthesia.
Pre‑operative variables
The analysis included the same set of variables for both of
the cohorts. Patient factors included age, gender, ASA clas-
sification and the primary diagnosis. The presence of a thick-
walled gallbladder (defined as ≥ 3mm or described as ‘thick
walled’) or CBD dilation (defined as a diameter of > 6mm)
on pre-operative ultrasound were also recorded, as well as
whether CT scans, MRCP or ERCP had been performed pre-
operatively [19]. The remaining factors related to the admis-
sion, namely whether the patient had previous admissions,
the type of admission (elective, delayed or emergency), and
the number of days from admission to surgery.
Nassar dicultly grading scale
In both datasets, surgeons were asked prospectively to grade
the difficulty of the procedure using the Nassar scale [17].
This scale graded operative findings from the gallbladder,
cystic pedicle and associated adhesions (Supplementary
Table1). The grading system is designed to be used as an

Surgical Endoscopy
1 3
overall summary of the operative conditions found; and
the worst factor found in the individual aspect of either the
‘Gallbladder’, ‘Cystic Pedicle’ or ‘Adhesions’ should be used
to define the final overall grade.
The scale was originally published in 1995, with grades
of 1–4, and subsequently modified in 1996 with the addition
of a grade 5. However, whilst the reference cohort used the
modified 5-point scale, the CholeS cohort was scored based
on the original 4-point scale. Since less than 1% of patients
in the reference database had a grade 5 operative difficulty,
grades 4 and 5 were combined, in order to make the two
cohorts comparable.
Statistical methods
Initially, univariable analyses were performed in the CholeS
cohort, to identify factors that were significantly associated
with difficult operations. For nominal variables, the rates of
difficult operations were compared across categories using
Chi-square tests. Ordinal variables were compared between
operations that were and were not classified as difficult using
Mann–Whitney tests.
The cohort was then randomly divided into two sets in
a 3:1 ratio, with the larger set used to derive a risk score.
A multivariable binary logistic regression model was pro-
duced, with a forwards stepwise approach used to select
factors that were independently associated with difficult
operations. The resulting model was then converted into a
risk score, the predictive accuracy of which was quantified
using ROC curves. The risk score was then calculated for
the smaller set of patients for internal validation, as well as
an external validation cohort, with the predictive accuracy
assessed using ROC curves.
All analyses were performed using IBM SPSS 22 (IBM
Corp. Armonk, NY), with p < 0.05 deemed to be indicative
of statistical significance throughout. Cases with missing
data for the difficulty score were excluded from analysis.
The multivariable analysis and risk score validation used a
complete cases approach, excluding patients with missing
data for any of the predictors considered.
Results
Patient demographics
Data were available for a total of 8755 operations in the
CholeS cohort, with the Nassar difficulty grading scale
recorded in 8680 (99.1%). Of these, 29.4% (N = 2548)
were of grade 3-5, and so classified as “difficult” (Fig.1).
Patients had a median age of 51years [interquartile range
(IQR) 38–64], and were predominantly female (74.0%) and
presented with biliary colic (55.1%). Further demographic
factors are reported in Table1.
Predictors ofdicult operations
On univariable analysis of the CholeS cohort (Table2),
operations were significantly more likely to be classified as
difficult with increasing age, ASA classification, time from
admission to surgery, as well as in males, those with thick-
walled gallbladders or CBD dilation, patients with previous
admissions, and those where pre-operative MRCP or ERCP
were performed. Significant differences across admission
types were also detected, with difficult operations least likely
in elective admissions. Rates of difficult operations also var-
ied with diagnosis, being highest in cholecystitis and lowest
in biliary colic.
Risk score derivation
The CholeS cohort was then divided into two separate sets
in a 3:1 ratio, with the larger set used for derivation of a risk
score, and the smaller set used for internal validation. The
derivation set was made up of N = 6507 operations with diffi-
culty scores recorded, of which 1929 (29.6%) were classified
as difficult. On multivariable analysis (Table3), increasing
age and ASA classification, male gender, diagnoses of CBD
stones or cholecystitis, thick-walled gallbladders, CBD dila-
tion, use of pre-operative ERCP and non-elective admissions
were all found to be significant independent predictors of
difficult operations.
The multivariable model was then converted into a risk
score. The odds ratios were first multiplied by two, to mini-
mise rounding errors, before being rounded to the nearest
integer, with two subtracted from the resulting values, such
that the lowest risk group for each factor would be assigned
a score of zero. In order to calculate the risk score for a
patient, the characteristics of the patient should be looked
up in Table4 for the eight factors, and the number of points
added up to produce a risk score with a potential range of
0–19. Within the CholeS derivation cohort, the median risk
score was 4 (IQR 2–7), with the highest observed score
being 17.
Fig. 1 Distribution of difficulty scores

Surgical Endoscopy
1 3
Internal validation
The risk score was then applied to the CholeS validation
cohort of 2173 operations, of which 619 (28.5%) were clas-
sified as difficult. The risk score was calculable for a total of
2064 operations, and returned an area under the ROC curve
(AUROC) for the prediction of difficult operations of 0.764
(95% CI 0.741–0.786, p < 0.001). The association between
the risk score and difficult operation rates across all three
cohorts is shown in Table5 and Fig.2.
Those patients within approximately the upper and
lower 25% of the distribution of the risk score were classi-
fied as high and low risk, respectively. As a result, patients
with risk scores of 0–1 (N = 426) were deemed low risk,
2–6 (N = 1034) as medium risk and 7+ (N = 604) as high
risk. Based on these classifications, the proportion of
patients with difficult operations was 8.7% (N = 37), 20.6%
(N = 213) and 55.3% (N = 334) for low, medium and high
risk, respectively.
External validation cohort
External validation was performed in the Reference cohort,
which was made up of 4089 operations, for which the Nas-
sar difficulty grading scale was recorded in 4035 (98.7%).
Of these, 35.1% (N = 1416) were classified as difficult
(grade 3–5), which was a significantly higher rate than for
the CholeS cohort (29.4%, p < 0.001, Fig.1). Comparisons
of demographics between the cohorts (Table1) found no
significant differences in age (p = 0.088) or ASA classi-
fication (p = 0.449). However, significant differences in
disease-related factors were observed, with the Reference
cohort having higher rates of biliary colic and lower rates
of thick-walled gallbladders (both p < 0.001). Patients in the
Reference cohort were also significantly less likely to have
MRCP or ERCP performed pre-operatively, and to be more
likely to be admitted as an emergency, due to the nature of
that service, with significantly longer times from admission
to surgery (all p < 0.001).
Table 1 Patient demographics
Data are reported as N (%), with p values from Ch-square tests, or as median (interquartile range), with p
values from Mann–Whitney tests, unless stated otherwise. Bold p values are significant at p < 0.05
*p value from Mann–Whitney test, as the factor is ordinal
CholeS Reference p value
NStatistic N Statistic
Age (years) 8673 51 (38–64) 3982 50 (38–63) 0.088
Gender (% female) 8680 6420 (74.0%) 4018 3050 (75.9%) 0.019
ASA classification 8609 3460 0.449*
1 3329 (38.7%) 1444 (41.7%)
2 4397 (51.1%) 1550 (44.8%)
3 861 (10.0%) 459 (13.3%)
4–5 22 (0.3%) 7 (0.2%)
Primary diagnosis 8675 4002 < 0.001
CBD stone 553 (6.4%) 587 (14.7%)
Cholecystitis 2511 (28.9%) 668 (16.7%)
Biliary colic 4777 (55.1%) 2479 (61.9%)
Pancreatitis 834 (9.6%) 268 (6.7%)
Thick-walled gallbladder 8476 2788 (32.9%) 4035 557 (13.8%) < 0.001
CBD dilation 8480 1343 (15.8%) 4035 631 (15.6%) 0.792
Previous admissions 8680 3626 (41.8%) 4035 943 (23.4%) < 0.001
Pre-Op. MRCP performed 8588 2246 (26.2%) 4035 171 (4.2%) < 0.001
Pre-Op. ERCP performed 8576 925 (10.8%) 4035 143 (3.5%) < 0.001
Admission type 8680 3982 < 0.001
Elective 4090 (47.1%) 1864 (46.8%)
Delay 3215 (37.0%) 854 (21.4%)
Emergency 1375 (15.8%) 1264 (31.7%)
Admission to surgery 8677 3910 < 0.001*
0days 7144 (82.3%) 1821 (46.6%)
1day 465 (5.4%) 808 (20.7%)
2days 305 (3.5%) 187 (4.8%)
3 + days 763 (8.8%) 1094 (28.0%)

Surgical Endoscopy
1 3
Table 2 Associations between
factors and difficult operations
in the CholeS and reference
cohorts
p values are from Ch-square tests, unless stated otherwise, and bold p values are significant at p < 0.05
*p value from Mann–Whitney test, as the factor is ordinal
CholeS Reference
NDifficult operations p Value NDifficult operations p Value
Age (years) < 0.001* < 0.001*
 < 40 2329 426 (18.3%) 1086 228 (21.0%)
40–49 1657 434 (26.2%) 816 252 (30.9%)
50–64 2581 851 (33.0%) 1189 456 (38.4%)
65+ 2106 835 (39.6%) 891 458 (51.4%)
Gender < 0.001 < 0.001
Female 6420 1549 (24.1%) 3050 912 (29.9%)
Male 2260 999 (44.2%) 968 499 (51.5%)
ASA classification < 0.001* < 0.001*
1 3329 719 (21.6%) 1444 361 (25.0%)
2 4397 1401 (31.9%) 1550 636 (41.0%)
3 861 392 (45.5%) 459 244 (53.2%)
4–5 22 16 (72.7%) 7 6 (85.7%)
Primary diagnosis < 0.001 < 0.001
CBD stone 553 229 (41.4%) 587 297 (50.6%)
Cholecystitis 2511 1302 (51.9%) 668 510 (76.3%)
Biliary colic 4777 789 (16.5%) 2479 496 (20.0%)
Pancreatitis 834 227 (27.2%) 268 101 (37.7%)
Thick-walled gallbladder < 0.001 < 0.001
No 5688 1136 (20.0%) 3478 981 (28.2%)
Yes 2788 1347 (48.3%) 557 435 (78.1%)
CBD Dilation < 0.001 < 0.001
No 7137 1922 (26.9%) 3404 1066 (31.3%)
Yes 1343 557 (41.5%) 631 350 (55.5%)
Previous admissions < 0.001 < 0.001
No 5054 1208 (23.9%) 3092 948 (30.7%)
Yes 3626 1340 (37.0%) 943 468 (49.6%)
Pre-Op. CT Performed < 0.001 < 0.001
No 7329 1969 (26.9%) 3972 1372 (34.5%)
Yes 1251 563 (45.0%) 63 44 (69.8%)
Pre-Op. MRCP performed < 0.001 < 0.001
No 6342 1735 (27.4%) 3864 1318 (34.1%)
Yes 2246 798 (35.5%) 171 98 (57.3%)
Pre-Op. ERCP performed < 0.001 < 0.001
No 7651 2108 (27.6%) 3892 1319 (33.9%)
Yes 925 422 (45.6%) 143 97 (67.8%)
Admission type < 0.001 < 0.001
Elective 4090 740 (18.1%) 1864 370 (19.8%)
Delay 3215 1145 (35.6%) 854 424 (49.6%)
Emergency 1375 663 (48.2%) 1264 603 (47.7%)
Admission to surgery < 0.001* < 0.001*
0 days 7144 1838 (25.7%) 1821 430 (23.6%)
1 days 465 203 (43.7%) 808 294 (36.4%)
2 days 305 158 (51.8%) 187 84 (44.9%)
3 + days 763 349 (45.7%) 1094 554 (50.6%)

Surgical Endoscopy
1 3
External validation oftherisk score
The risk score was calculable in 3340 operations from the
Reference cohort, with exclusions generally due to miss-
ing ASA classifications. Despite the differences in patient
characteristics between the cohorts, the predictive accu-
racy of the score in the Reference cohort was similar to that
previously described, with an AUROC of 0.789 (95% CI
0.773–0.806, p < 0.001). Using the previously defined risk
categories, the proportion of difficult operations was 11.0%
(78/712) in low risk, 31.1% (626/2012) in medium risk and
80.0% (493/616) in high risk patients.
Discussion
Intra-operative findings at the time of cholecystectomy vary
according to the clinical presentation, and may lead to a
range of operative challenges. The prediction of the diffi-
culty encountered during the procedure can offer the surgeon
a range of benefits, including surgical planning, informing
the patient, and predicting certain outcomes, such as the
potential for conversion to open surgery.
We found increasing age to be a significant risk factor in
predicting a difficult laparoscopic cholecystectomy. This was
similar to most published scoring systems [3, 4, 8, 9]. On the
other hand, some studies [10, 11, 21] found that age had no
significant correlation with difficulty when measured by the
conversion rate. This could be due to the small sample size
in these studies or, as reported by Mohanty etal. [10], due
to non-standardised experience of the operating surgeon. In
risk scores where age was dichotomized, a cut-off value of
50years was used in the majority of cases [3, 4]. However,
Table 3 Multivariable analysis of difficult operations on the CholeS
derivation cohort
Results are from a multivariable binary logistic regression model,
with difficult operation as the dependent variable. All factors from
Table 2 were considered for inclusion in the model, with variable
selection using a forwards stepwise approach. After exclusion of
patients with missing data, a total of N = 6188 (N = 1842 difficult
operations) were included in the final model. Bold p values are sig-
nificant at p < 0.05
OR (95% CI) p value
Age (Years) < 0.001
 < 40 – –
40–49 1.50 (1.23–1.82) < 0.001
50–64 1.68 (1.41–2.01) < 0.001
65+ 1.66 (1.37–2.01) < 0.001
Gender (male) 1.70 (1.49–1.95) < 0.001
ASA classification < 0.001
1 – –
2 1.32 (1.15–1.52) < 0.001
3 2.01 (1.62–2.50) < 0.001
4–5 4.32 (1.37–13.61) 0.012
Primary diagnosis < 0.001
Pancreatitis – –
Biliary Colic 1.23 (0.97–1.57) 0.092
CBD stone 1.40 (1.03–1.91) 0.031
Cholecystitis 2.91 (2.32–3.65) < 0.001
Thick-walled gallbladder ≥ 3mm 1.91 (1.67–2.20) < 0.001
CBD Dilation ≥ 6mm 1.36 (1.14–1.62) < 0.001
Pre-operative ERCP performed 1.60 (1.27–2.01) < 0.001
Admission type < 0.001
Elective – –
Delay 1.35 (1.15–1.58) < 0.001
Emergency 2.15 (1.77–2.61) < 0.001
Table 4 Risk score
Points
Age (years)
< 40 0
40+ 1
Gender
Female 0
Male 1
ASA Classification
1 0
2 1
3 2
4–5 7
Primary diagnosis
Pancreatitis 0
Biliary Colic 0
CBD stone 1
Cholecystitis 4
Thick-walled gallbladder (≥ 3mm)
No 0
Yes 2
CBD dilation (> 6mm)
No 0
Yes 1
Pre-operative ERCP
No 0
Yes 1
Admission type
Elective 0
Delayed 1
Emergency 2

Surgical Endoscopy
1 3
the multivariable analysis in this study found the risk of dif-
ficult operations to be similar in the 40–49, 50–64 and 65+
years groups, hence a cut-off of 40+ years was used in the
risk score.
The role of gender in relation to the disease process
resulting from gallstones has been explored by several pub-
lished studies, with male gender being a common predic-
tor of difficult cholecystectomy [3–5, 8, 12]. Yol etal. [22]
suggested that men with symptomatic gallstones are more
susceptible to inflammation and fibrosis.
ASA grade, whilst a marker of a patient’s general health
and fitness, was independently associated with worsening
difficulty in performing cholecystectomy and therefore
was included in our score. ASA is a globally recognised
score, which although is slightly subjective and has moder-
ate inter-rater reliability, it is well validated as a marker of
patients’ pre-operative health status [23]. Other papers have
also found the pre-operative ASA grade associated with a
variety of poor outcomes after cholecystectomy, including
conversion to open surgery [8, 9, 24], gangrenous gallblad-
der disease [25], worse post-operative complications and
need for post-operative procedures (for example IR drain-
age and ERCP) as well as length or stay, readmission and
post-operative mortality [26–28]. This is in keeping with
our analysis has shown a significant correlation between the
Table 5 Rates of difficult operations by risk score
Bold values highlight the totals in the subheadings of low, moderate and high risk patients
Risk score CholeS - derivation CholeS - validation Reference
Total NDifficult operations Total NDifficult operations Total NDifficult operations
0 343 27 (7.9%) 133 10 (7.5%) 257 30 (11.7%)
1 775 96 (12.4%) 293 27 (9.2%) 455 48 (10.5%)
Low risk subtotal 1118 123 (11.0%)426 37 (8.7%)712 78 (11.0%)
2 1153 164 (14.2%) 339 35 (10.3%) 688 130 (18.9%)
3 775 151 (19.5%) 270 62 (23.0%) 477 133 (27.9%)
4 536 139 (25.9%) 170 36 (21.2%) 393 129 (32.8%)
5 400 130 (32.5%) 146 37 (25.3%) 250 124 (49.6%)
6 370 129 (34.9%) 109 43 (39.4%) 204 110 (53.9%)
Medium risk subtotal 3234 713 (22.0%)1034 213 (20.6%)2012 626 (31.1%)
7 366 116 (31.7%) 120 56 (46.7%) 128 89 (69.5%)
8 408 187 (45.8%) 123 58 (47.2%) 124 82 (66.1%)
9 429 266 (62.0%) 141 75 (53.2%) 120 95 (79.2%)
10 386 243 (63.0%) 133 81 (60.9%) 136 128 (94.1%)
11 163 122 (74.8%) 62 41 (66.1%) 84 76 (90.5%)
12 70 55 (78.6%) 19 18 (94.7%) 19 18 (94.7%)
13 9 9 (100.0%) 4 3 (75.0%) 1 1 (100.0%)
14 2 1 (50.0%) 0 – 2 2 (100.0%)
15 0 – 1 1 (100.0%) 1 1 (100.0%)
16 2 2 (100.0%) 0 – 0 –
17 5 5 (100.0%) 1 1 (100.0%) 1 1 (100.0%)
High risk subtotal 1840 1006 (54.7%)604 334 (55.3%)616 493 (80.0%)
Fig. 2 Association between the risk score and rate of difficult opera-
tions. Points represent the proportion of operations classified as dif-
ficult for each value of the risk score. The final point combines
scores > 10, due to small numbers. The trendline is from a univariable
binary logistic regression model, with the risk score as a continuous
covariate, and using the pooled data from all three datasets

Surgical Endoscopy
1 3
ASA score and the difficulty of surgery, with rates of diffi-
cult operations of 22%, 32%, 46% and 73% for grades of 1,
2, 3 and 4–5, respectively in the CholeS cohort.
Siddiqui etal. [6], Lal etal. [15] and Carbotta etal. [16]
proposed pre- operative ultrasonography scoring systems to
predict difficulty. In our study, a thick-walled gallbladder
and a dilated CBD were significant predictors. A gallbladder
wall thickness over 3mm was suggested by some studies
[16, 29] while others [3, 4, 6, 9, 10, 14, 30] scored the gall
bladder wall when greater than 4mm. This corresponded to
the result in our study. Mudgal etal. [21] and Kumar A etal.
[31] included thickness of gallbladder wall in their scoring
systems, but did not specify the measurement. It was implied
that the thickness of the gallbladder wall causes difficulty in
grasping, manipulating and separating the gall bladder from
its bed, leading to a difficult procedure.
The CBD diameter was scored as a predictor of difficulty
by Siddiqui etal. [6], Vivek etal. [14] and Lal etal. [15], and
in our study, when it was > 6mm. However, Carmody etal.
[32] reported that pre- operative evaluation by ultrasound is
of little value in screening for difficult cases.
Pre-operative ERCP and endoscopic extraction of CBD
stones can increase the risk of pancreatitis and cause inflam-
matory changes in the CBD, which renders the dissection
of Calot’s triangle more difficult, thus making subsequent
cholecystectomy more challenging. Da Costa etal. [12] and
Ishizaki etal. [13] found pre-operative ERCP to be a strong
predictive factor for conversion to open surgery. This study
reflected these findings, confirming pre-operative ERCP as
a significant predictor of difficult surgery.
The variations in logistic issues and different hospital
protocols in the management of gallstone disease can lead
to some patients with a clinical diagnosis of acute chole-
cystitis undergoing delayed laparoscopic cholecystectomy.
Our study demonstrated that delayed cholecystectomy is a
significant predictor of difficulty. This is supported by stud-
ies including patients who had previous hospital admissions
secondary to gallstone disease [3–5, 7, 10, 12]. Bourgouin
etal. [5] propose clinical acute cholecystitis as a major pre-
dictor of difficult cholecystectomy. However, they consid-
ered the interval between the onset of symptoms and surgery
to be a better predictor of operative difficulty. The current
NICE guidelines for the management of cholecystitis suggest
performing laparoscopic cholecystectomy within 1week of
diagnosis to avoid difficult and challenging surgery [33].
There are no studies that included the type of admissions
in predictive scoring systems. However, Ashfaq etal. [11]
suggested the urgency of surgery as an independent risk fac-
tor for conversion to open. In our study, urgent admission
was found to be a significant independent predictor factor.
The other pre-operative predictive scoring systems pro-
posed in the literature used a range of different parameters to
define a difficult cholecystectomy. Many studies considered
conversion to open [3–6, 8–13] and long operating time
[3–7, 11, 12, 16] as measures of difficult laparoscopic chol-
ecystectomy. However, a decision of conversion to open sur-
gery and the operating time can vary greatly, depending on
the operator’s experience and skills. Other factors, such as
failure of access, failure of equipment and the need to deal
with vascular or bowel injuries, can influence the operating
time or the need to convert. These examples are independent
of any features of difficulty relating to the gallbladder or its
surrounding structures.
Onoe etal. [34] established a scoring system defining
difficulty by the ability to achieve the critical view of safety.
This may again be dependent on the surgeons’ skills and
whether they are using the critical view of safety to conclude
their dissection.
Lirici etal. [35] used the Nassar scale as an objective
assessment of difficulty to optimise the intra-operative man-
agement of complicated gallstone patients. We also clas-
sified difficulty according to the Nassar scale, which was
utilised and evaluated in the large multicentre prospective
CholeS study [18–20]. The dataset used for external vali-
dation was based on the practice of a single surgeon who
specialised in biliary surgery, and had performed more
than 4000 laparoscopic cholecystectomies over 20years,
while the CholeS study cohort included over 8000 surger-
ies performed by multiple surgeons with different levels of
experience. When used in the CholeS study, the operative
difficulty grade reporting may have been subject to some
variability between the large numbers of participating sur-
geons, although they were given clear definitions in the pro-
tocol and encouraged to access example laparoscopic videos
detailing different operative difficulty grades. The objective
measure of operative difficultly assessment, rather than the
varied and subjective measures used in other scores is a
major benefit of our scoring system. Table6 compares data
from some previously published operative difficultly scores
and our series. Other scores with small numbers of patients
or very few risk factors were excluded.
Our predictive score is based on simple and easily obtain-
able variables, which would be readily available to the sur-
geon pre-operatively. It was derived using data from a large
prospective series, and validated in an external cohort.
Comparisons between these cohorts identified some base-
line clinical differences between them. The reference cohort
had a significantly higher incidence of emergency admis-
sions, due to the nature of the service, which was designed
to accept the great majority of emergency biliary admissions
and to deliver index admission cholecystectomy. It also had
a significant percentage of patients undergoing bile duct
explorations. However, despite these differences, validation
of the risk score returned similar results for both datasets,
a testament to the predictive score’s clinical usefulness and
applicability in a variety of scenarios.

Surgical Endoscopy
1 3
Table 6 Comparison of selected pre-operative difficultly scores for use in laparoscopic cholecystectomy
Definition of operative difficulty Difficulty factors/score
Gupta etal. [3]; Prospective; internal validation; N = 210
Time taken > 60min
Bile/stone spillage
Injury to duct
Conversion to open
Age > 50years
Male sex
History of hospitalisation for acute cholecystitis
BMI > 25
Abdominal scar
Palpable gallbladder
Wall thickness ≥ 4mm
Pericholecystic collection
Impacted stone
Randhawa etal. [4]; Retrospective; internal validation; N = 228
Difficult:
Not converted to open, but one of the following:
Time taken 60–120min
Bile/stone spillage
Injury to duct
Very difficult:
Time taken > 120min
Conversion to open
Age > 50
Male sex
History of hospitalisation due to acute cholecystitis
BMI > 25
Abdominal scar (infra umbilical or supra umbilical)
Palpable gallbladder
Wall thickness ≥ 4mm
Pericholecystic collection
Impacted stone
Bourgouin etal. [5]; Retrospective; internal validation; N = 420
Operative time ≥ 1.5 times the surgeon’s individual base time
Procedures converted to open surgery
Male sex
Previous cholecystitis attack
Fibrinogen count
Neutrophil count
Alkaline phosphatase count
Siddiqui etal [6]; Retrospective; no validation; N = 300
Time taken > 60min
Presence of biliary leakage
Injury to duct/artery
Conversion to open
Gallbladder wall thickness ≥ 4mm
Transverse diameter of gallbladder ≥ 5cm
Presence of impacted stones
CBD diameter > 6mm
Presence of pericholecystic collection
Number of stones > 1
Liver size ≥ 15.5cm
Schrenk etal. [7]; Prospective; internal validation; N = 640
Degree of adhesion to the gallbladder
Presence of scarring in the triangle of Calot
Acute or chronic inflammatory changes
Duration of surgery > 60min
WBC > 10× 109/l
No visualisation of gallbladder
Either thickened gallbladder wall > 5mm, hydroptic
gallbladder, pericholecystic fluid
Shrunken gallbladder
Previous upper abdominal surgery
Biliary colic within the last 3weeks
Right upper quadrant pain
Rigidity in right upper abdomen
Kanakala etal. [8]; Retrospective and prospective; internal validation; N = 2117
Conversion to open Male sex
ASA Classification
Rosen etal. [9]; Retrospective; Internal Validation; N = 1347
Conversion to open procedure Increasing age
BMI > 30 for AC and BMI > 40 for elective cholecystitis
Acute cholecystitis
Gallbladder wall thickness > 4mm
ASA > 2 in non-elective cholecystectomies
Mohanty etal. [10]; Retrospective; internal validation; N = 512
Conversion to open Number of attacks > 5
Total leucocyte count > 11,000/cu mm
Gallbladder wall thickness > 4mm

Surgical Endoscopy
1 3
Conclusion
This study has developed an internally and externally
validated pre-operative scoring system for predicting the
operative difficulty of laparoscopic cholecystectomy. This
scoring system was validated using objective measures of
operative difficulty, as opposed to the largely subjective
measures used in previous studies. Our score, employ-
ing simple pre-operative variables, can accurately predict
the likelihood of a difficult operation, facilitating patient
selection for day case surgery, optimising pre-operative
surgical planning and helping to inform patients during
the consent process. We could also imagine surgeons using
the scoring system to decide on whether a specialist or
senior gallbladder surgeon may be required in helping
with the case if a high level of difficultly is predicted.
This could allow difficult operations to be scheduled for
the daytime in the correct theatres and with the correct
surgical expertise and equipment. It could also be used to
optimise training for different grades of surgical trainees
and ensure that they get appropriate training cases for their
level of surgical skill. The score could also be used to risk
adjust outcomes in future research in this area.
Acknowledgements
CholeS Study Management Team
Ravinder S. Vohra, Consultant Surgeon, Nottingham Oesophago-
Gastric Unit, Nottingham University Hospitals NHS Foundation Trust,
Hucknall Road, Nottingham, UK;
Amanda J. Kirkham, Biostatistician; Cancer Research UK Clinical
Trials Unit, The University of Birmingham, Birmingham, UK;
Sandro Pasquali, Surgical trainee, Surgical Oncology Unit, Veneto
Institute of Oncology IOV-IRCCS, Padova, Italy;
Paul Marriott, Previous Surgical trainee, West Midlands Research
Collaborative, Academic Department of Surgery, The University of
Birmingham, Birmingham, UK;
Marianne Johnstone, Previous Surgical trainee, West Midlands
Research Collaborative, Academic Department of Surgery, The Uni-
versity of Birmingham, Birmingham, UK;
Philip Spreadborough, Previous Surgical trainee, West Midlands
Research Collaborative, Academic Department of Surgery, The Uni-
versity of Birmingham, Birmingham, UK;
Derek Alderson, Emeritus Professor of Surgery, Academic Depart-
ment of Surgery, The University of Birmingham, Birmingham, UK;
Ewen A. Griffiths, Consultant Surgeon, Department of Upper Gas-
trointestinal Surgery, University Hospitals Birmingham NHS Founda-
tion Trust, Birmingham, UK
CholeS Study Collaborators:
England—Stephen Fenwick, Mohamed Elmasry, Quentin M.
Nunes, David Kennedy (Aintree University Hospital NHS Foundation
Trust); Raja Basit Khan, Muhammad A. S. Khan (Airedale General
Hospital); Conor J. Magee, Steven M. Jones, Denise Mason, Ciny P.
Risk factors in bold-italic are the same as, or analogous to, factors included in the score for the present study ±
Table 6 (continued)
Definition of operative difficulty Difficulty factors/score
Ashfaq etal. [11]; Retrospective; Internal Validation; N = 351
Conversion to open
Operative time > 120 AND one of the following:
“Necrotic, gangrenous or perforated” GB,
Extensive lysis of adhesions,
Prior cholecystostomy tube insertion
Mirizzi syndrome
Urgency of operation
Previous abdominal surgery
Severe inflammation/gangrene
DW da Costa etal. [12]; Prospective; Internal Validation; N = 249
Visual analogue scale score ≥ 8
Duration of surgery > 75min
Conversion or subtotal cholecystectomy
Male sex
Prior sphincterotomy
Delayed cholecystectomy for at least 2weeks
Ishizaki etal. [13]; Retrospective; internal validation; N = 1179
Conversion to open procedure Thickened gallbladder wall > 4mm
History of CBD stones removed by pre-operative ERCP
Present study; prospective; internal + external validation; N = 8755 + 4089
Objective intra-operative difficulty (Nassar scale) Age (40 + years)
Increasing ASA Classification
Male sex
Primary diagnosis of cholecystitis or CBD stone
Thick-walled gallbladder > 4mm
CBD dilation > 6mm
Use of pre-operative ERCP
Non-elective operations

Surgical Endoscopy
1 3
Parappally (Wirral University Teaching Hospital); Pawan Mathur,
Michael Saunders, Sara Jamel, Samer Ul Haque, Sara Zafar (Barnet
and Chase Farm Hospital); Muhammad Hanif Shiwani, Nehemiah
Samuel, Farooq Dar, Andrew Jackson (Barnsley District General Hos-
pital); Bryony Lovett, Shiva Dindyal, Hannah Winter, Ted Fletcher,
Saquib Rahman (Basildon Univesity Hospital); Kevin Wheatley, Tom
Nieto, Soofiyah Ayaani (Sandwell and West Birmingham Hospitals
NHS Trust); Haney Youssef, Rajwinder S. Nijjar, Helen Watkin, David
Naumann, Sophie Emesih; Piyush B. Sarmah, Kathryn Lee, Nikita Joji,
Joel Lambert (Heart of England Foundation NHS Trust); Jonathan
Heath, Rebecca L. Teasdale, Chamindri Weerasinghe (Blackpool
Teaching Hospitals NHS Foundation Trust); Paul J. Needham, Hannah
Welbourn, Luke Forster, David Finch (Bradford Teaching Hospitals
NHS Foundation Trust); Jane M. Blazeby, William Robb, Angus G. K.
McNair, Alex Hrycaiczuk (University Hospitals Bristol NHS Trust);
Alexandros Charalabopoulos, Sritharan Kadirkamanathan, Cheuk-
Bong Tang, Naga V. G. Jayanthi, Nigel Noor (Broomfield Hospital);
Brian Dobbins, Andrew J. Cockbain, April Nilsen-Nunn, Jonathan de
Siqueira (Calderdale and Huddersfield NHS Trust); Mike Pellen, Jona-
than B. Cowley, Wei-Min Ho, Victor Miu (Hull and East Yorkshire
NHS Trust); Timothy J. White, Kathryn A. Hodgkins, Alison Kinghorn
(Chesterfield Royal Hospital NHS Foundation Trust); Matthew G. Tut-
ton, Yahya A. Al-Abed, Donald Menzies, Anwar Ahmad, Joanna Reed,
Shabuddin Khan (Colchester Hospital University NHS Foundation
Trust); David Monk, Louis J. Vitone, Ghulam Murtaza, Abraham Joel
(Countess of Chester NHS Foundation Trust); Stephen Brennan, David
Shier, Catherine Zhang, Thusidaran Yoganathan (Croydon Health Ser-
vices NHS Trust); Steven J. Robinson, Iain J. D. McCallum, Michael
J. Jones, Mohammed Elsayed, Liz Tuck, John Wayman, Kate Carney
(North Cumbria University Hospitals Trust); Somaiah Aroori, Kenneth
B. Hosie, Adam Kimble, David M. Bunting, Kenneth B. Hosie (Plym-
outh Hospitals NHS Trust); Adeshina S. Fawole, Mohammed Basheer,
Rajiv V. Dave, Janahan Sarveswaran, Elinor Jones, Chris Kendal (Mid
Yorkshire NHS Trust); Michael P. Tilston, Martin Gough, Tom Wal-
lace, Shailendra Singh, Justine Downing Katherine A. Mockford, Eyad
Issa, Nayab Shah, Neal Chauhan (Northern Lincolnshire and Goole
NHS Foundation Trust); Timothy R. Wilson, Amir Forouzanfar, Jona-
than R. L. Wild, Emma Nofal, Catherine Bunnell, Khaliel Madbak
(Doncaster and Bassetlaw Hospitals NHS Foundation Trust); Sudhin-
dra T. V. Rao, Laurence Devoto, Najaf Siddiqi, Zechan Khawaja (Dor-
set County Hospital NHS Foundation Trust); James C. Hewes, Laura
Gould, Alice Chambers, Daniel Urriza Rodriguez (North Bristol NHS
Trust); Gourab Sen, Stuart Robinson, Kate Carney, Francis Bartlett
(Freeman Hospital); David M. Rae, Thomas E. J. Stevenson, Kas Sar-
vananthan (Frimley Park Hospital NHS Trust); Simon J. Dwerryhouse,
Simon M. Higgs, Oliver J. Old, Thomas J. Hardy, Reena Shah Steve T.
Hornby, Ken Keogh, Lucinda Frank (Gloucestershire Hospitals NHS
Trust); Musallam Al-Akash, Emma A. Upchurch (Great Western Hos-
pitals NHS Foundation Trust); Richard J. Frame, Michael Hughes,
Clare Jelley (Harrogate and District NHS Foundation Trust); Simon
Weaver, Sudipta Roy, Toritseju O. Sillo, Giorgios Galanopoulos (Wye
Valley NHS Trust); Tamzin Cuming, Pedro Cunha, Salim Tayeh, Sar-
antos Kaptanis (Homerton University Hospital NHS Trust); Mohamed
Heshaishi, Abdalla Eisawi, Michael Abayomi; Wee Sing Ngu, Katie
Fleming, Dalvir S. Bajwa (Tees Hospitals NHS Foundation Trust);
Vivek Chitre, Kamal Aryal, Paul Ferris (Paget University Hospitals
NHS Foundation Trust); Michael Silva, Simon Lammy Sarah
Mohamed, Amir Khawaja, Adnan Hussain, Mudassar A. Ghazanfar,
Maria Irene Bellini (Oxford University NHS Trust); Hamdi Ebdewi,
Mohamed Elshaer, Gianpiero Gravante, Benjamin Drake (Kettering
General Hospital NHS Foundation Trust); Arikoge Ogedegbe,
Dipankar Mukherjee, Chanpreet Arhi, Lola Giwa Nusrat Iqbal (Bark-
ing, Havering and Redbridge University Hospitals NHS Trust); Nicho-
las F. Watson, Smeer Kumar Aggarwal, Philippa Orchard, Eduardo
Villatoro (Kings Mill Hospital); Peter D. Willson, Kam Wa Jessica
Mok, Thomas Woodman, Jean Deguara (Kingston Hospital NHS
Foundation Trust); Giuseppe Garcea, Benoy I. Babu, A. R. Dennison,
Deep Malde, David Lloyd, Steve Satheesan, Omer Al-Taan, Alexander
Boddy (University Hospitals of Leicester NHS Trust); John P. Slavin,
Robert P. Jones, Laura Ballance, Stratos Gerakopoulos (Leighton Hos-
pital, Mid Cheshire Hospitals NHS Foundation Trust); Periyathambi
Jambulingam, Sami Mansour, Naomi Sakai, Vikas Acharya (Luton &
Dunstable University Hospital NHS Foundation Trust); Mohammed
M. Sadat, Lawen Karim, David Larkin, Khalid Amin (Macclesfield
District General Hospital); Amarah Khan, Jennifer Law, Saurabh Jam-
dar, Stella R. Smith, Keerthika Sampat, Kathryn M. O’shea (Central
Manchester NHS Foundation Trust); Mangta Manu, Fotini M. Asprou,
Nabeela S. Malik, Jessica Chang, Marianne Johnstone (Royal Wolver-
hampton Hospitals NHS Trust); Michael Lewis, Geoffrey P. Roberts,
Babu Karavadra, Evangelos Photi (Norfolk and Norwich University
Hospitals NHS Foundation Trust); James Hewes, Laura Gould, Alice
Chambers, Dan Rodriguez (North Bristol NHS Trust); Derek A.
O’Reilly, Anthony J. Rate, Hema Sekhar, Lucy T. Henderson, Benja-
min Z. Starmer, Peter O. Coe, Sotonye Tolofari, Jenifer Barrie (Pennine
Acute NHS Trust); Gareth Bashir, Jake Sloane, Suroosh Madanipour,
Constantine Halkias, Alexander E. J. Trevatt (North Middlesex Trust);
David W. Borowski, Jane Hornsby, Michael J. Courtney, Suvi Virupak-
sha (North Tees and Hartlepool NHS Foundation Trust); Keith Sey-
mour, Sarah Robinson, Helen Hawkins, Sadiq Bawa, Paul V. Gallagher,
Alistair Reid, Peter Wood (Northumbria Healthcare NHS Foundation
Trust); J. G. Finch, J. Guy Finch, J. Parmar, E. Stirland (Northampton
General Hospital NHS Trust); James Gardner-Thorpe, Ahmed Al-
Muhktar, Mark Peterson, Ali Majeed (Sheffield Teaching Hospitals
NHS Foundation Trust); Farrukh M. Bajwa, Jack Martin, Alfred Choy,
Andrew Tsang (Peterborough City Hospital); Naresh Pore, David R.
Andrew, Waleed Al-Khyatt, Christopher Taylor, Santosh Bhandari,
Adam Chambers, Dhivya Subramanium (United Lincolnshire Hospitals
NHS Trust); Simon K. C. Toh, Nicholas C. Carter, Sophie Tate,
Belinda Pearce, Denise Wainwright, Stuart J. Mercer, Benjamin Knight
(Portsmouth Hospitals NHS Trust); Vardhini Vijay, Swethan Alagarat-
nam, Sidhartha Sinha, Shahab Khan (The Princess Alexandra Hospital
NHS Trust); Shamsi S. El-Hasani, Abdulzahra A. Hussain (Kings Col-
lege Hospital NHS Foundation Trust); Vish Bhattacharya, Nisheeth
Kansal, Tani Fasih, Claire Jackson (Gateshead Health NHS Foundation
Trust); Midhat N. Siddiqui, Imran A. Chishti, Imogen J. Fordham,
Zohaib Siddiqui (Lewisham and Greenwich NHS Trust); Harald Baus-
bacher, Ileana Geogloma, Kabita Gurung (Queen Elizabeth Hospital
NHS Trust); George Tsavellas, Pradeep Basynat, Ashish Kiran
Shrestha, Sanjoy Basu, Alok Chhabra Mohan Harilingam, Mohamed
Rabie, Mansoor Akhtar (East Kent Hospitals University NHS Founda-
tion Trust); Pradeep Kumar, Sadaf F. Jafferbhoy, Najam Hussain, Sou-
lat Raza (Burton Hospitals NHS Foundation Trust); Manzarul Haque,
Imran Alam, Rabiya Aseem, Shakira Patel, Mehek Asad (Royal Albert
Edward Infirmary, Wigan Wrightington and Leigh NHS Trust);
Michael I. Booth, William R. Ball, Christopher P. J. Wood, Ana C.
Pinho-Gomes (Royal Berkshire NHS Foundation Trust); Ambareen
Kausar, Moh’d Rami Obeidallah (East Lancashire Hospital Trust);
Joseph Varghase, Joshil Lodhia, Donal Bradley, Carla Rengifo, David
Lindsay (Royal Bolton Hospital NHS Foundation Trust); Sivakumar
Gopalswamy, Ian Finlay, Stacy Wardle, Naomi Bullen (Royal Cornwall
NHS Trust); Syed Yusuf Iftikhar, Altaf Awan, Javed Ahmed, Paul
Leeder (Royal Derby NHS Foundation Trust); Guiseppe Fusai, Giles
Bond-Smith, Alicja Psica, Yogesh Puri (Royal Free, London); David
Hou, Fergus Noble, Karoly Szentpali, Jack Broadhurst (Hampshire
Hospital NHS Foundation Trust); Ravindra Date, Martin R. Hossack,
Yan Li Goh, Paul Turner, Vinutha Shetty (Lancashire Teaching Hos-
pitalsNHS Foundation Trust); Manel Riera, Christina A. W. Macano,
Anisha Sukha (Royal Shrewsbury Hospital); Shaun R. Preston, Jennifer
R. Hoban, Daniel J. Puntis, Sophie V. Williams (Royal Surrey County
Hospital NHS Foundation Trust); Richard Krysztopik, James Kynas-
ton, Jeremy Batt, Matthew Doe (Royal United Hospital Bath NHS
Trust); Andrzej Goscimski, Gareth H. Jones, Stella R. Smith, Claire

Surgical Endoscopy
1 3
Hall (Salford Royal NHS Foundation Trust); Nick Carty, Jamil Ahmed,
Sofoklis Panteleimonitis (Salisbury Hospital Foundation Trust); Rohan
T. Gunasekera, Andrea R. G. Sheel, Hannah Lennon, Caroline Hindley
(Southport and Ormskirk Hospital NHS Trust); Marcus Reddy, Ross
Kenny, Natalie Elkheir, Emma R. McGlone (St George’s Healthcare
NHS Trust); Rajasundaram Rajaganeshan, Kate Hancorn, Anita Har-
greaves (St Helens and Knowsley Teaching Hospitals NHS Trust); Raj
Prasad, David A. Longbotham, Dhakshinamoorthy Vijayanand, Imeshi
Wijetunga (Leeds Teaching Hospitals); Paul Ziprin, Christopher R.
Nicolay, Geoffrey Yeldham, Edward Read (Imperial College Health-
care NHS Trust); James A. Gossage, Rachel C. Rolph, Husam Ebied,
Manraj Phull (St Thomas’ Hospital, London); Mohammad A. Khan,
Matthew Popplewell, Dimitrios Kyriakidis, Anwar Hussain (Mid Staf-
fordshire NHS Foundation Trust); Natasha Henley, Jessica R. Packer,
Laura Derbyshire, Jonathan Porter (Stockport NHS Foundation Trust);
Shaun Appleton, Marwan Farouk, Melvinder Basra (Bucks Healthcare
NHS Trust); Neil A. Jennings, Shahda Ali, Venkatesh Kanakala (City
Hospitals Sunderland NHS Foundation Trust); Haythem Ali, Risha
Lane, Richard Dickson-Lowe, Prizzi Zarsadias (Tunbridge Wells and
Maidstone NHS Trust); Darius Mirza, Sonia Puig, Khalid Al Amari,
Deepak Vijayan, Robert Sutcliffe, Ravi Marudanayagam (University
Hospital Birmingham NHS Foundation Trust); Zayed Hamady,
Abheesh R. Prasad, Abhilasha Patel (University Hospital Coventry and
Warwickshire NHS Trust); Damien Durkin, Parminder Kaur, Laura
Bowen (University Hospital of North Staffordshire NHS Trust); James
P. Byrne, Katherine L. Pearson, Theo G. Delisle, James Davies (Uni-
versity Hospital Southampton NHS Foundation Trust); Mark A. Tom-
linson, Michelle A. Johnpulle, Corinna Slawinski (University Hospitals
of Morecambe Bay); Andrew Macdonald, James Nicholson, Katy
Newton, James Mbuvi (University Hospital South Manchester NHS
Foundation Trust); Ansar Farooq, Bhavani Sidhartha Mothe, Zakhi
Zafrani, Daniel Brett (Warrington and Halton Hospitals NHS Trust);
James Francombe, Philip Spreadborough, James Barnes, Melanie
Cheung (South Warwickshire NHS Foundation Trust); Ahmed Z. Al-
Bahrani, Giuseppe Preziosi, Tomas Urbonas (Watford General Hospi-
tal); Justin Alberts, Mekhlola Mallik, Krashna Patel, Ashvina Segaran,
Triantafyllos Doulias (West Suffolk NHS Trust); Pratik A. Sufi, Caro-
line Yao, Sarah Pollock (Whittington NHS Trust); Antonio Manzelli,
Saj Wajed, Michail Kourkulos, Roberto Pezzuto (Wonford Hospital);
Martin Wadley, Emma Hamilton, Shameen Jaunoo, Robert Padwick
(Worcestershire Acute Hospitals NHS Trust); Mazin Sayegh, Richard
C. Newton, Madhusoodhana Hebbar, Sameh F. Farag, (Western Sussex
Hospitals NHS Foundation Trust); John Spearman, Mohammed F.
Hamdan, Conrad D’Costa, Christine Blane; (Yeovil District Hospital
NHS Trust); Mathew Giles, Mark B. Peter, Natalie A. Hirst, Tanvir
Hossain, Arslan Pannu Yesar El-Dhuwaib, Tamsin E. M. Morrison,
Greg W. Taylor (York Teaching Hospital NHS Foundation Trust).
Northern Ireland—Ronald L. E. Thompson, Ken McCune, Paula
Loughlin, Roger Lawther (Altnagelvin Area Hospital); Colman K.
Byrnes, Duncan J. Simpson, Abi Mawhinney, Conor Warren (Antrim
Area Hospital); Damian McKay, Colin McIlmunn, Serena Martin, Mat-
thew MacArtney (Daisy Hill Hospital); Tom Diamond, Phil Davey,
Claire Jones, Joshua M. Clements, Ruairi Digney, Wei Ming Chan, Ste-
phen McCain, Sadaf Gull, Adam Janeczko, Emmet Dorrian, Andrew
Harris, Suzanne Dawson, Dorothy Johnston, Barry McAree, (Belfast
City Hospital, Mater Infirmorum Hospital Belfast and Royal Victoria
Hospital); Essam Ghareeb, George Thomas, Martin Connelly, Ste-
phen McKenzie, Krzysztos Cieplucha (South West Acute Hospital);
Gary Spence, William Campbell, Gareth Hooks, Neil Bradley (Ulster
Hospital).
Republic of Ireland—Arnold D. K. Hill, John T. Cassidy, Michael
Boland (Beaumont Hospital, Dublin); Paul Burke, Deirdre M. Nally
(University Hospital Limerick); Arnold D. K. Hill, Elmoataz Khogali,
Wael Shabo, Edrin Iskandar (Louth County Hospital and Our Lady of
Lourdes Hospital); Gerry P. McEntee, Maeve A. O’Neill, Colin Pei-
rce, Emma M. Lyons (Mater Hospital, Dublin); Adrian W. O’Sullivan,
Rohan Thakkar, Paul Carroll, Ivan Ivanovski (Mercy University Hospi-
tal); Paul Balfe, Matthew Lee (St Luke’s General Hospital Kilkenny);
Des C. Winter, Michael E. Kelly, Emir Hoti, Donal Maguire; Priya-
darssini Karunakaran, Justin G. Geoghegan, Frank McDermott, Sean T.
Martin (St Vincent’s University and Private Hospitals, Dublin); Keith
S. Cross, Fiachra Cooke, Saquib Zeeshan, James O. Murphy (Waterford
Regional Hospital); Ken Mealy, Helen M. Mohan, Yuwaraja Nedujche-
lyn, Muhammad Fahad Ullah (Wexford General Hospital).
Scotland—Irfan Ahmed, Francesco Giovinazzo, James Milburn
(Aberdeen Royal Infirmary); Sarah Prince, Eleanor Brooke, Joanna
Buchan (Belford Hospital); Ahmed M. Khalil, Elizabeth M. Vaughan,
Michael I. Ramage, Roland C. Aldridge (Borders General Hospital);
Simon Gibson, Gary A. Nicholson, David G. Vass (Crosshouse Hospi-
tal, Ayrshire & Arran); Alan J. Grant, David J. Holroyd, M. Angharad
Jones, Cherith MLR Sutton (Dr Gray’s Hospital); Patrick O’Dwyer,
Frida Nilsson (Gartnavel General Hospital); Beatrix Weber, Tracey
K. Williamson, Kushik Lalla, Alice Bryant (Gilbert Bain Hospital);
C. Ross Carter, Craig R. Forrest, David I. Hunter (Glasgow Royal
Infirmary); Ahmad H. Nassar, Mavis N. Orizu, Katrina Knight,
Haitham Qandeel (Monklands Hospital); Stuart Suttie, Rowena Beld-
ing, Andrew McClarey (Ninewells Hospital); Alan T. Boyd, Graeme
J. K. Guthrie, Pei J. Lim, Andreas Luhmann (Perth Royal Infirmary);
Angus J. M. Watson, Colin H. Richards, Laura Nicol, Marta Madur-
ska (Raigmore Hospital); Ewen Harrison, Kathryn M. Boyce, Amanda
Roebuck, Graeme Ferguson (Royal Infirmary of Edinburgh); Pradeep
Pati, Michael S. J. Wilson, Faith Dalgaty, Laura Fothergill (Stracathro
Hospital); Peter J. Driscoll, Kirsty L. Mozolowski, Victoria Banwell,
Stephen P. Bennett (Victoria Hospital, Kirkcaldy); Paul N. Rogers,
Brendan L. Skelly, Claire L. Rutherford, Ahmed K. Mirza (Western
Infirmary Glasgow).
Wales—Taha Lazim, Henry C. C. Lim, Diana Duke, Talat Ahmed
(Bronglais General Hospital); William D. Beasley, Marc D. Wilkinson,
Geta Maharaj, Cathy Malcolm (Glangwili General and Prince Philip
Hospital); Timothy H. Brown, Bilal Al-Sarireh, Guy M. Shingler,
Nicholas Mowbray, Rami Radwan (Morriston and Singleton Hospi-
tals); Paul Morcous, Simon Wood, Abbas Kadhim (Princess of Wales
Hospital); Duncan J. Stewart, Andrew L. Baker, Nicola Tanner, Hri-
shikesh Shenoy (Wrexham Maelor Hospital).
Data validators—Shazia Hafiz, Joshua A. De Marchi, Deepak
Singh-Ranger, Elzanati Hisham, Paul Ainley, Stephen O’Neill, John
Terrace, Sara Napetti, Benjamin Hopwood, Thomas Rhys, Justine
Downing, Osama Kanavati, Maria Coats, Danail Aleksandrov, Char-
lotte Kallaway, Salama Yahya, Beatrix Weber, Alexa Templeton, Mar-
tin Trotter, Christina Lo, Ajit Dhillon, Nick Heywood, Yousif Aawsaj,
Alhafidz Hamdan, Obuobi Reece-Bolton, Andrew McGuigan, Yousef
Shahin, Aymon, Ali Alison Luther, James A. Nicholson, Ilayaraja
Rajendran, Matthew Boal, Judith Ritchie.
Funding None.
Compliance with ethical standards
Disclosures Drs. Nassar, Ng, Vohra, Katbeh, Zino, and Griffiths and
Mr Hodson have no conflict of interest or financial ties to disclose.
References
1. Salky B, Edye M (1998) The difficult cholecystectomy: problems
related to concomitant diseases. Semin Laparosc Surg 5:107–114
2. Laws H (1998) The difficult cholecystectomy: problems during
dissection and extraction. Semin Laparosc Surg 5:81–91

Surgical Endoscopy
1 3
3. Gupta N, Ranjan G, Arora M (2013) Validationof a scoring sys-
tem to predict difficult laparoscopic cholecystectomy. Int J Surg
11(9):1002–1006
4. Randhawa J, Pujahari A (2009) Preoperative prediction of difficult
lap chole: a scoring method. Indian J Surg 71(4):198–201
5. Bourgouin S, Mancini J, Monchal T (2016) How topredict dif-
ficult laparoscopic cholecystectomy? Proposal for a simple pre-
operative scoring system. Am J Surg 212(5):873–881
6. Siddiqui M, Rizvi S, Sartaj S, Ahmad I, Rizvi S (2017) A stand-
ardized ultrasound scoring system for preoperative prediction
of difficult laparoscopic cholecystectomy. J Med Ultrasound
25:227–231
7. Schrenk P, Woisetschlager R, Rieger R, Wayand W (1998) A diag-
nostic score to predict the difficulty of a laparoscopic cholecystec-
tomy from preoperative variables. Surg Endosc 12:148–150
8. Kanakala V, Borowski D, Pellen M, Dronamraju S, Woocock S,
Seymour K etal (2011) Risk factors in laparoscopic cholectstomy:
a multivariate analysis. Int J Surg 9(4):318–323
9. Rosen M, Brody F, Ponsky J (2002) Predictive factorsfor conver-
sion of laparoscopic cholecystectomy. Am J Surg 183(3):254–258
10. Mohanty S, Mohanty R (2017) Pre-operative prediction of difficult
laparoscopic cholecystectomy using clinical and ultrasonographic
parameters. Ann Int Med Dent Res 3(4):43–48
11. Ashfaq A, Ahmadieh K, Shah A, Chapital A, Harold K, Johnson D
(2016) The difficult gall bladder: outcomes following laparoscopic
cholecystectomy and the need for open conversion. Am J Surg
212:1261–1264
12. da Costa DW, Schepers NJ, Bouwense SA, Hollemans RA, van
Santvoort HC, Bollen TL etal (2018) Predicting a ‘difficult chole-
cystectomy’ after mild gallstone pancreatitis. HPB 21(7):827–833
13. Ishizaki Y, Miwa K, Yoshimoto J, Sugo H, Kawasaki S (2006)
Conversion of elective laparoscopic to open cholecystectomy
between 1993 and 2004. Br J Surg 93:987–991
14. Vivek M, Augustine A, Rao R (2014) A comprehensive predic-
tive scoring method for difficult laparoscopic cholecystectomy. J
Minim Access Surg 10(2):62–67
15. Lal P, Agarwal P, Malik V, Chakravati A (2002) A difficult laparo-
scopic that requires conversion to open procedure can be predicted
by preoperative ultrasonography. JSLS 6:59–63
16. Carbotta G, Panebianco A, Laforgia R, Pascazio B, Balducci G,
Bianchi F etal (2018) A new clinical ultrasound score to predict
difficult videolaparocholecystectomies: a prospective study. Ann
Med Surg 35:59–63
17. Nassar A, Ashkar K, Mohamed A, Hafiz A (1995) Is laparoscopic
cholecystectomy possible without video technology? Minim Inva-
sive Ther 4(2):63–65
18. Griffiths E, Hodson J, Vohra R, Marriott P, Katbeh T, Zino S
etal (2019) Utilisation of an operative difficulty grading scale for
laparoscopic cholecystectomy. Surg Endosc 33:110–121
19. Vohra R, Spreadborough P, Johnstone M, Marriott P, Bhangu A,
Alderson D etal (2015) Protocol for a multicentre, prospective,
population-based cohort study of variation in practice of cholecys-
tectomy and surgical outcomes (The CholeS study). BMJ Open
5:e006399
20. CholeS Study Group WMRC (2016) Population-based cohort
study of outcomes following cholecystectomy for benign gall-
bladder diseases. Br J Surg 103(12):1704–1715
21. Mudgal M, Kushwah N, Singh R, Gehlot H (2018) A clinical
study to determine predictive factors for difficult laparoscopic
cholecystectomy. Int J Med Sci Public Health 7(2):116–121
22. Yol S, Kartal A, Vatansev C, Aksoy F, Toy H (2006) Sex as a
factor in conversion from laparoscopic cholecystectomy to open
surgery. JSLS 10:359–363
23. Sankar A, Johnson S, Beattie W, Tait G, Wijeysundera D (2014)
Reliability of the American Society of Anaesthesiologists physical
status scale in clinical practice. Br J Anaesth 113(3):424–432
24. Sutcliffe R, Hollyman M, Hodson J, Bonney G, Vohra R, Griffiths
E etal (2016) Pre- operative risk factors for conversion from lapa-
roscopic to open cholecystectomy: a validated risk score derived
from a prospective UK database of 8820 patients. HPB (Oxford)
18(11):922–928
25. Wu B, Buddensick T, Ferdosi H, Narducci D, Sautter A, Setiawan
L etal (2014) Predicting gangrenous cholecystitis. HPB (Oxford)
16(9):801–806
26. Harboe K, Bardram L (2011) The quality of cholecystectomy in
Denmark: outcome and risk factors for 20307 patients from the
national database. Surg Endosc 25(5):1630–1641
27. Jensen K, Roth N, Krarup P, Bardram L (2019) Surgical manage-
ment of acute cholecystitis in a nationwide Danish cohort. Lan-
genbecks Arch Surg 404(5):589–597
28. Giger U, Michel J, Opitz I, Inderbitzin DT, Kocher T, Krahen-
buhl L etal (2006) Risk factors for perioperative complications
in patients undergoing laparoscopic cholecystectomy: analysis of
22953 consecutive cases from the Swiss Association of Lapa-
roscopic and Thoracoscopic Surgery database. J Am Coll Surg
203(5):723–728
29. Acharya A, Adhikari S (2012) Preoperative scoring system to
predict difficult laparoscopic cholecystectomy. PMJN 12(1):45–50
30. Agrawal N, Singh S, Khichy S (2015) Preoperative prediction of
difficult laparoscopic cholecystectomy: a scoring method. Niger
J Surg 21(2):130–133
31. Kumar A, Singh S, Chhabra A, Nemma S (2017) Assessment of
the efficacy of a score based system used for assigning various
specific scores for predicting difficult laparosopic cholecystec-
tomy. Int J Contemp Med Res 4(4):861–864
32. Carmody E, Arenson A, Hanna S (1994) Failed or difficult lapa-
roscopic cholecystectomy: can preoperative ultrasonography iden-
tify potential problems. J Clin Ultrasound 22(6):391–396
33. National Institute for Health and Care Excellence (2014) Gallstone
disease. Diagnosis and management. NICE, London
34. Onoe S, Maeda A, Takayama Y, Fukami Y, Kaneoka Y (2017)
A preoperative predictive scoring system to predict the ability to
achieve the critical view of safety during laparoscopic cholecys-
tectomy for acute cholecystitis. HPB 19:406–410
35. Lirici M, Califano A (2010) Management of complicated gall-
stones: results of an alternative approach to difficult cholecystec-
tomies. Minim Invasive Ther Allied Technol 19(5):304–315
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.