Surgery in Practice and Science 14 (2023) 100211

Contents lists available at ScienceDirect

Surgery in Practice and Science

journal homepage: www.sciencedirect.com/journal/surgery-in-practice-and-science

Whipple’s procedure for pancreatic cancer: training and the hospital

environment are more important than volume alone
Shamir O. Cawich a, *, Robyn Cabral a, Jacintha Douglas a, Dexter A. Thomas a,
Fawwaz Z. Mohammed a, Vijay Naraynsingh a, Neil W. Pearce b
a
Department of Clinical Surgical Sciences, University of the West Indies, St Augustine Campus, Trinidad & Tobago, West Indies
b
Department of Surgery, Southampton University NHS Trust, Southampton, United Kingdom SO16DP

A R T I C L E I N F O A B S T R A C T

Keywords: Background: In our center, patients with pancreatic cancer traditionally had Whipple’s resections by general
Pancreas surgery teams until January 2013 when a hepatopancreatobiliary (HPB) was introduced. We compared outcomes
Whipples before and after introduction of HPB teams.
Resection
Methods: Data were collected from the records of all patients booked for Whipple’s resections over a 12-year
Outcomes
period. The data were divided into two groups: Group A consisted of the 6-year period from January 1, 2007
Volume
to December 30, 2012 during which all resections were performed by GS teams. Group B comprised patients in
the 6-year period from January 1, 2013 to December 30, 2019 during which operations were performed by HPB
teams. All statistical analyses were carried out using SPSS ver 16.0 and a P Value <0.05 was considered sta­
tistically significant.
Results: The patients selected for Whipple’s resections in Group A had statistically better performance status and
lower anaesthetic risk. Despite this, patients in Group A had higher conversions to palliative operations (66% vs
5.3%), longer mean operating time (517±25 vs 367±54 min; P<0.0001), higher blood loss (3687±661 vs 1394
±656 ml; P<0.0001), greater transfusion requirements (4.3±1.3 vs 1.9±1.4 units; P<0.001), greater likelihood
of prolonged ICU stay (100% vs 40%; P=0.19), higher overall morbidity (75% vs 22.2%; P=0.02), higher major
morbidity (75% vs 13.9%; P=0.013), more procedure-related complications (75% vs 9.7%; P=0.003) and higher
mortality rates (75% vs 5.6%; P<0.0001). The HPB teams were more likely to perform vein resection and
reconstruction to achieve clear margins (26.4% vs 0; P=0.57).
Conclusion: This paper adds to the growing body of evidence that volume alone should not be used as a marker of
quality for patients requiring Whipple’s procedures.

Introduction all pancreatic resections were performed by General Surgery (GS) teams.
We sought to determine whether there were any differences in short
In the English-Speaking Caribbean, pancreatic cancer occurs at an term outcomes when pancreatic head adenocarcinomas were treated by
incidence of 4-4.5 per 100,000 persons per annum [1]. It is accepted that HPB vs GS teams.
a Whipple’s procedure is the best therapeutic option for adenocarci­
noma at the head of the pancreas [2]. Methods
Although it is a complex operation, the Whipple’s procedure is
considered safe when carried out by trained hepatopancreatobiliary This Caribbean nation has a population of 1.4 million persons [6].
(HPB) teams in specialized centers [2–3]. Over the past decade there has Under the auspices of the Caribbean Chapter of the Americas Hep­
been a trend to refer these patients to centralized hospitals for HPB atopancreatobiliary Association (AHPBA) an attempt was made at ser­
teams to perform Whipple’s operations in large volumes [4–5]. vice centralization for hepatobiliary diseases with the establishment of
At our institution in Trinidad & Tobago, a small island state in the three HPB units in 2011 [6]. The General Hospital in Port of Spain
Eastern Caribbean, HPB services were introduced in 2013. Prior to this, houses one of these HPB units [6]. At this facility, a formal HPB unit was

* Corresponding author.
E-mail address: tt.liver.surgery@gmail.com (S.O. Cawich).

https://doi.org/10.1016/j.sipas.2023.100211
Received 3 July 2023; Received in revised form 14 August 2023; Accepted 15 August 2023
Available online 15 August 2023
2666-2620/© 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
S.O. Cawich et al. Surgery in Practice and Science 14 (2023) 100211

incorporated in January 2013. This HPB unit was comprised of 2 In Group A, there were 9 patients with potentially resectable
fellowship-trained HPB surgeons, one specialized anaesthetist, a dedi­ pancreatic head adenocarcinomas who were booked for Whipple’s
cated senior registrar and 2 junior residents. All cases were discussed in procedures. There were 4 males and 5 females at a mean age of 54.7
multidisciplinary team meetings where therapeutic decisions were years (SD ±7.5; Range 44-65; Median 55). In this group, 5 (56%) pa­
made. Prior to incorporation of the HPB team, all liver and pancreatic tients were booked for Whipple’s procedures but deemed to have irre­
resections at this institution were performed by GS teams. sectable disease at surgical exploration. In these patients, the operative
We secured institutional review board approval to collect data from procedures were converted to surgical palliative bypasses. These pa­
all patients who underwent Whipple’s procedures at this facility. We tients were excluded from further analyses. Four (44%) patients had
sought to collect data for a period of 12 years, from January 1, 2007 to completion of the planned Whipple’s procedure. In this group, the mean
December 30, 2019. This period was deliberately chosen to compare age was 50.8 years (SD ±7.4; range 44-59; Median 50). Tables 1 and 2
short-term outcomes before and after the January 2013 date when the document the ECOG scores and ASA scores in this group.
HPB service was incorporated. In Group B, 76 patients with pancreatic head carcinomas were
We retrospectively collected data from the registers at the pathology booked for Whipple’s procedures. There were 34 males and 42 females
department, operating theatres, ICU department and MDT records to at a mean age of 60.2 years (SD±9.28; range 46-77; Median 61). In this
identify all consecutive patients with pancreatic head adenocarcinomas group, 4 (5.3%) patients were deemed irresectable at the time of oper­
during the study period. The subset of patients who were booked for ation and underwent palliative bypasses. These patients were excluded
Whipple’s resections were identified and their hospital records were from the final sample of 72 patients who had completion of Whipple’s
retrieved for detailed review. procedures. Tables 1 and 2 document the ASA and ECOG scores in this
We excluded patients who were not candidates for Whipple’s re­ group.
sections, those whose paper-based records could not be retrieved, pa­
tients who underwent intra-operative conversions to palliative bypass Operative details
procedures and those who underwent laparoscopic Whipple’s
operations. In Group A, the mean operating time for Whipple’s procedures was
The following data were extracted: patient demographics, perfor­ 517.5 minutes (Range 490-550; SD±25; Median 515). The operations in
mance status using the Eastern Cooperative Oncology Group (ECOG) these patients were accompanied by a mean blood loss of 3687.5 ml
classification, physical status using the American Society of Anaes­ (Range 2500-5000; SD± 661.44; Median 3500) and mean packed cell
thesiologists (ASA) scoring system, estimated operative blood loss, transfusion requirements of 4.25 units (Range 3-6; SD ± 1.26; Median
duration of operation (from incision to closure), post-operative com­ 4). There were no vascular resections or reconstructions performed in
plications, 30-day mortality, completion of Whipple’s operations, con­ this group.
versions to surgical palliation, duration of hospitalization and In Group B, the mean operating time was 367 minutes (Range 260-
transfusion requirements. 485; SD±54.1; Median 350). The operations in these patients were
Complications were classified using the modified Clavien-Dindo accompanied by a mean blood loss of 1394 ml (Range 600-4000; SD
system [7]. Pancreatic leak was categorized according to the Interna­ ±656.8; Median 1200) and mean transfusion requirements of 1.88 units
tional Study Group on Pancreatic Fistula (ISGPF) criteria [8]. Massive of packed cells (Range 0-5; SD ±1.43; Median 2). Nineteen (26.4%)
operative bleeding was defined as the loss of >1 blood volume within patients underwent planned resections and reconstruction of the supe­
the operative period or 50% of the patient’s blood volume in <3 hours rior mesenteric/portal vein. Operative details are compared in Table 3.
[9]. Massive transfusion was defined as receipt of >10 units of packed
red cells in 24 hours or >6 units within 6 hours [10].
Complications were also subdivided into medical and procedure- Hospitalization details
related complications. Medical complications include aspirations,
pneumonia, pulmonary failure, deep vein thromboses, pulmonary In this setting, we maintained a policy of mandatory ICU admission
embolus, myocardial infarction, arrhythmias, congestive heart failure, after Whipple’s resection because institutional limitations generally did
renal failure and septicemia [11–12]. Procedure-related complications not allow the expected level of supportive care outside of the ICU setting.
include massive intra-operative bleeding, pancreatic fistula, delayed In Group A, the mean ICU stay was 12 days (Range 6-12; SD±2.65;
gastric emptying, surgical site infection, organ space collection, pseu­ Median 8). All (100%) patients required a prolonged ICU stay >72 hours
doaneurysm and post-operative haemorrhage [11–12]. for invasive treatment, ventilator and/or inotropic support. One patient
The records of all patients who were booked for Whipple’s resections in this group survived to discharge after 17 days of hospitalization.
were retrieved and data were divided into two groups: Group A con­ In Group B, the mean ICU stay of 5.24 days (Range 1-40; SD±7.22;
sisted of the 6-year period from January 1, 2007 to December 30, 2012 Median 3). Twenty-nine (40.3%) patients required a prolonged ICU stay
during which all resections were performed by GS teams. Group B beyond 72 hours for invasive treatment, ventilator and/or inotropic
consisted of patients in the 6-year period from January 1, 2013 to support. Overall, the mean duration of hospitalization after Whipple’s
December 30, 2019 during which operations were performed by HPB procedure was 15.1 days (Range 8-60; SD±9.53; Median 12). The results
teams. are compared in Table 4.
All data were compared between the two groups. The t-test for in­
dependent means was used to compare continuous numerical values Table 1
between the groups. The Chi Square and test were used to compare ASA Scores for Patients Undergoing Whipple’s Procedures.
categorical variables between the groups. Statistical analyses were car­ Score ASA Descriptor GS HPB P
ried out using SPSS ver 16.0 and a P Value <0.05 was considered sta­ I Completely healthy 4 10 0.0006
tistically significant. (100%) (13.9%)
II Mild Systemic Disease 0 24 -
(33.3%)
Results
III Severe Systemic Disease, not 0 30 -
incapacitating (41.7%)
Over the 12-year study period, 85 patients were diagnosed with IV Incapacitating disease that is a threat 0 8 (11.1%) -
potentially operable pancreatic head carcinomas. Data from patients to life
with pancreatic adenocarcinoma who were not candidates for Whipple’s V Moribund and not expected to 0 0 -
survive > 24 Hours
procedures were excluded.

2
S.O. Cawich et al. Surgery in Practice and Science 14 (2023) 100211

Table 2 Table 5
Performance Scores for Patients Undergoing Whipple’s Procedures. Complications after Whipple’s Procedures.
Grade ECOG Performance Status GS HPB P Morbidity Description GenSx HPB P

Overall Number of patients with any 3 16 0.018

0 Fully active, able to carry out all 4 13 0.0015 complication (75%) (22.2%)
activities without restriction (100%) (18.1%) Minor Clavien-Dindo I or II 1 6 (8.3%) 0.262
1 Restricted in physically strenuous 0 20 - (25%)
activity, but ambulatory and able to (27.8%) • Pneumonia 1 2
carry out light work • Deep Vein Thrombosis 0 1
2 Ambulatory and capable of self-care, 0 34 - • Delayed gastric emptying 0 1
but unable to carry out work (47.2%) • Gastrointestinal bleeding 0 2
activities. Up and about >50% of Major Clavien-Dindo III or IV 3 10 0.013245
waking hours (75%) (13.9%)
3 Capable of limited self-care and 0 4 (5.6%) - • Anastomotic dehiscence 0 1
confined to bed or chair for more than • Massive upper GI Bleeding 0 1
50% of waking hours • Myocardial infarction 1 3
4 Completely disabled and cannot carry 0 1 (1.4%) - • Pseudoaneurysm 1 2
on self-care. Confined to bed or chair • Biliary sepsis as a source of 0 2
5 Dead 0 0 - septicemia
• POPF / organ space 0 1
collection
• Massive Operative Bleeding 1 0
Table 3 Mortality 30-day mortality 3 4 (5.6%) <0.00001
Operative Details in Patients Undergoing Whipple’s Resections. (75%)

Parameter GS HPB P

Number of patients booked for PD 11 76 outlined in Table 6.

Conversions to palliative bypasses 7 (63.6%) 4 (5.26%) <0.0001
Completed Whipple’s resections 4 (36.4%) 72
Discussion
<0.0001
(94.7%)
Planned vascular resections / 0 19 0.5667
reconstruction (26.4%) Whipple’s procedure is the only existing treatment with a potential
Mean operating time in mins (mean 517.5±25 367±54.1 <0.0001 to cure pancreatic head malignancies [13–14]. Fortuitously, modern
t-value
±SD)
series have documented improved complication profiles, with modern
5.51092
Estimated blood loss in mls (Mean 3687.5 1394 <0.0001
30-day mortality rates between 4-6% [13–17].
±SD) ±661.44 ±656.8 t-value Many factors have been credited as contributors to the improved
6.60718 morbidity profile, including a multidisciplinary approach to care [2–3],
Transfusion requirements in units 4.25±1.26 1.88±1.43 0.000894 advanced cross-sectional imaging [2,18], specialized surgical equip­
(Mean ±SD) t-value
ment [6], ICU availability [19–20], readily available support services
3.24056
[17–20] and quaternary surgical training [21]. Prior to the introduction
of a HPB service at our facility, advanced cross-sectional imaging, ICU
and support services were available. After the introduction of the HPB
Table 4
service in January 2013, a multidisciplinary HPB team was established
Hospitalization in Patients Undergoing Whipple’s Resections.
and met weekly to discuss cases. Specialized surgical equipment was
Parameter GS HPB P also procured. Although the center remained a low volume center [6],
I.C.U admission in days (Mean ±SD) 12± 5.24±7.22 0.326408 considerable effort was invested to develop hospital-specific protocols
2.65 t-value geared to the delivery of subspeciality care. We also developed a
0.45169
modified centralization concept that was tailored to our low-volume
Prolonged ICU stay >72 Hours (Mean 4 29 0.01878
SD) (100%) (40.3%)
environment, described in a previous publication [22]. There was
Total hospital stay in days (Mean 17 * 15.1±9.53 - good stakeholder buy-in at our institution, and patients requiring
±SD) Whipple’s procedures were referred to the HPB team. This study has
*
Calculation inappropriate with a single value. shown that there were demonstrable benefits as a result of these
changes.
Morbidity / mortality analysis
Case volumes
Three (75%) patients experienced at least one complication in Group
A. Major complications occurred in 3 (75%) patients, all resulting in There was a significant increase in case volumes from 0.7 to 12
death within 30-days of operation. These deaths were due to shock
secondary to massive intra-operative haemorrhage (1), myocardial Table 6
infarction (1) and respiratory failure secondary a pneumonia (1). Procedure-Related Complications after Whipple’s Procedures.
In Group B, 16 (22.2%) patients experienced at least one complica­ Description GenSx HPB P
tion. Minor complications were recorded in 6 (8.3%) patients and major (4) (72)
complications in 10 (14.0%) patients. The individual complications are Procedure-related complications only 3 (75%) 7 0.002705
outlined in Table 5. There were 4 (5.5%) deaths in Group B, due to (9.7%)
massive bleeding from a pseudoaneurysm that could not be controlled at • Post-operative pancreatic fistula / organ 0 1
space collection
re-operation (1), septic shock secondary to biliary sepsis (1), anasto­
• Massive intra-operative bleeding 1 0
motic leak with intra-abdominal sepsis and multiple organ failure (1) • Massive transfusion requirement 1 2
and a massive myocardial infarction (1). • Delayed gastric emptying 0 1
When medical complications were excluded, significantly more pa­ • Pseudoaneurysm 1 2
tients in Group A experienced procedure-related complications as • Anastomotic leak 0 1

3
S.O. Cawich et al. Surgery in Practice and Science 14 (2023) 100211

Whipple’s procedures per annum after introduction of the HPB team. stay for invasive treatment / ventilator support in the HPB group (40.3%
This was likely multifactorial due to (1) changes in referral patterns with vs 100%). It was difficult to compare overall hospitalization, since only
increased numbers of patients referred to the service and (2) HPB sur­ one patient survived to discharge after 17 days in Group A. These data
gical teams having higher thresholds to convert to surgical palliation suggest that the patients, having undergone smoother operations, had
because they were more comfortable with complex resections. less requirement for post-operative support.
Although the case volume increased, we still do not consider this a
high-volume center. There is no consensus on the definition of a high- Complication profile
volume center, but most researchers quote numbers >18 Whipple’s
procedures per annum as high volume [3,13,23–29] and this facility The results have shown that there was a significant reduction in the
performed a mean of 12 resections annually. risk of overall complications (22% vs 75%), major morbidity (14% vs
75%) and mortality (5.6% vs 75%) when Whipple’s resections were
Resection rates performed by HPB teams. These results are in keeping with existing data
that demonstrated an inverse relationship between case-volumes and
There was a statistically significant increase in completed resections overall morbidity [2,3,23] plus 30-day mortality [2,3,23,30–32]
by HPB surgeons (94.7% vs 36.4%). Considering that resection is the following Whipple’s resections.
only option that brings a chance for cure, patients with pancreatic Many authorities have suggested that there is a difference between
cancer are being better served by the HPB teams with greater thresholds medical complications and procedure specific complications [11,12].
for conversion to palliative procedures. We believe this was a reflection Medical complications include aspirations, pneumonia, pulmonary
of HPB surgeons being more experienced and willing to perform more failure, renal failure and septicemia [11,12]. Procedure-related com­
complex maneuvers, such as portal vein resections, to achieve R0 plications include pancreatic fistula, delayed gastric emptying, deep
resections. surgical site infections, pseudoaneurysms and intra-abdominal hae­
morrhage [11]. In our series procedure-specific complications were
Patient complexity significantly lower in Group B (9.7% vs 50%), and this is reflective of
technical expertise. While surgical expertise is necessary, it is insuffi­
It was interesting that the patients undergoing resections by GS cient to guarantee good post-operative outcomes [3,11,12]. Medical
teams were highly selected. Most teams would operate on patients with complications were also greater in group A (50% vs 13.2%), reflective
ASA scores ≤II and ECOG scores ≤I, but in this study, there was a sig­ the need for improvement in pre-habilitation.
nificant disparity, with all patients in group A having ASA scores I In Group B, there were experienced surgeons and team development
(100% vs 14%) and ECOG scores 0 (100% vs 18.1). It appeared that GS to account for the improved outcomes. However, it is important to note
teams had a significantly lower threshold to declare patients medically that this is a resource-poor center, with a paucity of blood products, ICU
unfit to undergo Whipple’s resections. space, operating lists and specialized equipment. This brings special
The HPB teams were willing to attempt resections in sicker patients. meaning to the fact that good outcomes can be achieved in these settings
In this group, 53% had ASA scores ≥3 and 54% had ECOG performance and outcomes cannot be based on volume data alone.
scores ≥2. This is an important point as patients with pancreatic carci­ Recently, there has been a move away from simple volume data to­
noma tend to present late and are more likely to be diagnosed when their ward examining surgeon / team experience using several surrogate
decompensation has commenced. markers that include: surgeon experience which was defined as one who
had performed >50 Whipple’s procedures in their career [3]; technical
Operative complexity competence which can be defined by the proportion of vein resection /
reconstruction as a surrogate marker [3] and a surpassed learning curve,
Schmidt et al. [3] suggested that the performance of vein resection / defined by Tseng et al [33] as >60 Whipple’s procedures. By all metrics,
reconstruction can be used as a surrogate marker of technical complexity the HPB teams in this facility have achieved and surpassed the metrics.
as well as surgeon experience. In this series, more vascular re­ Experience is important for such a technically complex operation.
constructions were performed by HPB teams (26.4% vs 0). This The experienced surgeon would know how to resect / reconstruct the
approached, but did not achieve statistical significance, but it is a clin­ portal vein when required to achieve negative margins [3], when not to
ically important distinction. We agree with Schmidt et al. [3] that it is a operate on patients [34], to recognize aberrant anatomy [34], how to
complex maneuver, but it is important that pancreatic surgeons are get out of trouble when complications occur intra-operatively [13].
willing and able to perform this maneuver when required to achieve a We also believe that two additional factors must be taken into ac­
curative resection. count. Firstly, the surgeon must be willing to adapt to a new working
Schmidt et al. [3] also introduced the concept of the “experienced environment with numerous limitations, such as unavailability of ICU
surgeon” which they defined as one who had performed >50 Whipple’s space, operating lists, paucity of blood supplies, etc. It is important to
procedures in their career. The surgeons leading the HPB teams in the adapt practice to focus on peri-operative management and inter-
Caribbean were experienced surgeons having accrued sufficient expe­ disciplinary cooperation that evolved with time. This interaction and
rience in their career to perform >50 Whipple’s procedures. It was not continuous learning are not limited to the surgical team alone. Instead, it
possible for us to compare this metric as there were no statistics avail­ involves continuous, adaptive learning by the entire institution [1,2,7,
able for GS teams. 19] by the development of prehabilitation, multidisciplinary team
Considering that HPB surgeons performed more complex resections interaction, intra-operative anaesthesia care, surgeon training,
in patients who were sicker, it was an unexpected finding that operative post-operative care pathways, post-procedure nursing care, ICU care,
time, estimated blood loss and transfusion requirements were all availability of emergency medical doctors and experienced subspeciality
significantly greater in Group A. This points to a difference in technical supportive care [2,3,13,23,35–37].
facility between the operating teams.
Limitations
Post-operative support services
The retrospective study design inherently introduced a limitation in
In this study, there were trends toward shorter duration of ICU stay data collection. Also, because historical data were used retrospectively,
(5.24 vs 12 days) when HPB teams performed resections, and there was it was not possible to evaluate parameters such as training, surgeon
a statistically significant reduction in the likelihood of prolonged ICU experience, etc.

4
S.O. Cawich et al. Surgery in Practice and Science 14 (2023) 100211

Conclusion [13] van Heek NT, Kuhlmann KFD, Scholten RJ, de Castro SM, Busch ROC, van
Gulik TM, Obertop H, Gouma DJ. Hospital Volume and Mortality After Pancreatic
Resection: A Systematic Review and an Evaluation of Intervention in The
This paper adds to the growing body of evidence that volume alone Netherlands. Ann Surg 2005;242(6):781–90.
should not be used as a marker of quality for patients requiring Whip­ [14] Kuhlmann KF, de Castro SM, Wesseling JG, et al. Surgical treatment of pancreatic
ple’s procedures. Despite low volumes at our facility, we demonstrated adenocarcinoma; actual survival and prognostic factors in 343 patients. Eur J
Cancer 2004;40:549–58.
that all metrics improved when specialty teams were introduced, [15] Kotwall CA, Maxwell JG, Brinker CC, Koch GG, Covington DL. National estimates
hospital-based protocols were developed and continuous adaptive of mortality rates for radical pancreaticoduodenectomy in 25,000 patients. Ann
learning by the entire hospital were observed. These data support a Surg Oncol 2002;9(9):847–54.
[16] Ho V, Heslin MJ. Effect of hospital volume and experience on in-hospital mortality
move away from the traditional volume concept. for pancreaticoduodenectomy. Ann Surg 2003;237(4):509–14.
[17] Billimora, et al. Comparison of Perioperative Mortality and Long-Term Survival for
Funding Cancer Surgery. J Clin Oncol 2008;26:4624–33.
[18] Mayo SC, Gilson MM, Herman JM, Cameron JL, Nathan H, Edil BH, et al.
Management of patients with pancreatic adenocarcinoma: national trends in
This research did not receive any specific grant from funding patient selection, operative management, and use of adjuvant therapy. J Am Coll
agencies in the public, commercial, or not-for-profit sectors. Surg 2012;214(1):33–45.
[19] Simianu VV, Zyromski NJ, Nakeeb A, Lillemoe KD. Pancreatic cancer: progress
made. Acta Oncol 2010;49(4):407–17.
Disclosure statement [20] Ljungqvist LO, Dejong CH, Demartines N, Parks RW, Lobo DN, et al.
Pancreaticoduodenectomy: ERAS recommendations. Clin Nutr 2013;32(5):870–1.
[21] Francis W, Arra A, Bonadie KO, Cawich SO. Evolution of Liver and Pancreas
There was no funding made available to support this work.
Surgical Sub-Specialty in the Caribbean: Caribbean Chapter of the Americans
Hepatopancreatobiliary Association. J Carib Coll Surg 2021;1:51–4.
Declaration of Competing Interest [22] Cawich SO, Pearce NW, Naraynsingh V, Shukla P, Deshpande RR. Whipple’s
operation with a modified centralization concept: A model in low-volume
Caribbean centers. World J Clin Cases 2022;10(22):7620–30.
The authors declare that they have no known competing financial [23] Bliss LA, Yang CJ, Chau Z, Ng SC, McFadden DW, Kent TS, Moser AJ, Callery MP,
interests or personal relationships that could have appeared to influence Tseng JF. Patient selection and the volume effect in pancreatic surgery: unequal
the work reported in this paper. benefits? HPB 2014;16:899–906.
[24] McPhee JT, Hill JS, Whalen GF, Zayaruzny M, Litwin DE, Sullivan ME, et al.
Perioperative mortality for pancreatectomy: a national perspective. Ann Surg
References 2007;246:246–53.
[25] Alsfasser G, Kittner J, Eisold S, Klar E. Volume-outcome relationship in pancreatic
[1] Rawla P, Sunkara T, Gaduputic V. Epidemiology of Pancreatic Cancer: Global surgery: the situation in Germany. Surgery 2012;152(3S1):50–5.
Trends, Etiology and Risk Factors. World J Oncol 2019;10(1):10–27. [26] Glasgow RE, Mulvihill SJ. Hospital Volume Influences Outcome in Patients
[2] Soreide JA, Sandvik OM, Soreide K. Improving pancreas surgery over time: undergoing pancreatic resection for cancer. West J Med 1996;165(5):294–300.
Performance factors related to transition of care and patient volume. Int J Surg [27] Sosa JA, Bowman HM, Gordon TA, et al. Importance of hospital volume in the
2016;32:116–22. overall management of pancreatic cancer. Ann Surg 1998;228(3):429–38.
[3] Schmidt CM, Turrini O, Parikh P, House MG, Zyromski NJ, Nakeeb A, Howard TJ, [28] Meguid, et al. What constitutes a ‘high-volume’ hospital for pancreatic resection?
Pitt HA, Lillemoe KD. Effect of Hospital Volume, Surgeon Experience, and Surgeon J Am Coll Surg 2008;206:622–9.
Volume on Patient Outcomes After Pancreaticoduodenectomy: A Single-Institution [29] Briceno P, Hutson J, Shridhar R, Meredit K. Pancreatic Resection at High Volume
Experience. Arch Surg 2010;145(7):634–40. Centers Improves Survival. HPB 2017;S171(19):131. https://doi.org/10.1016/j.
[4] Fong Y, Gonen M, Rubin D, Radzyner M, Brennan MF. Long-Term Survival Is hpb.2017.02.384.
Superior After Resection for Cancer in High-Volume Centers. Ann Surg 2005;242: [30] Gordon TA, Bowman HM, Tielsch JM, et al. Statewide regionalization of
540–7. pancreaticoduodenectomy and its effect on in-hospital mortality. Ann Surg 1998;
[5] Neoptolemos JP, Russell RC, Bramhall S, et al. Low mortality following resection 228:71–8.
for pancreatic and periampullary tumours in 1026 patients: UK survey of specialist [31] Derogar M, Blomberg J, Sadr-Azodi O. Hospital teaching status and volume related
pancreatic units. UK Pancreatic Cancer Group. Br J Surg 1997;84:1370–6. to mortality after pancreatic cancer surgery in a national cohort. BJS 2015;102:
[6] Cawich SO, Kluger MD, Francis W, Deshpande RR, Mohammed F, Bonadie KO, 548–57. https://doi.org/10.1002/bjs.9754.
DA Thomas, Pearce NW, Schrope BA. Review of minimally invasive pancreas [32] Birkmeyer JD, Warshaw AL, Finlayson SR, Grove MR, Tosteson AN. Relationship
surgery and opinion on its incorporation into low volume and resource poor between hospital volume and late survival after pancreaticoduodenectomy. Surg
centres. World J Gastrointest Surg 2021;13(10):1122–35. https://doi.org/ 1999;126(2):178–83.
10.4240/wjgs.v13.i10.1122. [33] Tseng JF, Pisters PW, Lee JE, et al. The learning curve in pancreatic surgery.
[7] Teoule P, Bartel F, Birgin E, Ruckert F, Wilhelm TJ. The Clavien-Dindo Surgery 2007;141(4):456–63.
Classification in Pancreatic Surgery: A Clinical and Economic Validation. J Invest [34] Bouvet M. Comment on the effect of Hospital Volume, Surgeon Experience, and
Surg 2018;16:1–7. Surgeon Volume on Patient Outcomes After Pancreaticoduodenectomy: A Single-
[8] Bassi C, Marchegiani G, Dervenis C, Sarr M, Abu-Hilal M, et al. The 2016 update of Institution Experience. Arch Surg 2010;145(7):640.
the International Study Group (ISGPS) definition and grading of postoperative [35] Gasper WJ, Glidden DV, Jin C, Way LW, Patti MG. Has recognition of the
pancreatic fistula: 11 Years After. Surgery 2017;161(3):584–91. relationship between mortality rates and hospital volume for major cancer surgery
[9] Mavrides E, Allard S, Chandraharan E, et al. Prevention and Management of in California made a difference? A follow-up analysis of another decade. Ann Surg
Postpartum haemorrhage. BJOG 2016;16(5):e106–49. 124. 2009;250(3):472–83.
[10] Zatta AJ, Mcquilten ZK, Mitra B. Elucidating the clinical characteristics of patients [36] Hashimoto DA, Bababekov Y, Mehtsun WT, Stapleton SM, Warshaw AL,
captured using different definitions of massive transfusion. Vox Sang 2014;107(1): Lillemoe KD, et al. Is Annual Volume Enough? The Role of Experience and
60–70. Specialization on Inpatient Mortality After Hepatectomy. Ann Surg. 2017;266(4):
[11] Ho CK, Kleef J, Friess H, Buchler MW. Complications of Pancreatic Surgery. HPB 603–9.
2005;7(2):99–108. https://doi.org/10.1080/13651820510028936. [37] Ihse I. The volume-outcome relationship in cancer surgery: a hard sell. Ann Surg
[12] Halloran CM, Ghaneh P, Bosonnet L, Hartley MN, Sutton R, Neoptolemos JP. 2003;238:777–81.
Complications of pancreatic cancer resection. Dig Surg 2002;19:138–46.