Abstract
Background
For logistical reasons, some high-volume centers have developed surgical programs wherein 1 surgical team performs 2 pancreatoduodenectomies on a single day. It is unclear whether this practice has a negative impact on surgical outcome.
Methods
We conuducted a retrospective analysis including all consecutive open pancreatoduodenectomies in a single high-volume center (2014–2021). Pancreatoduodenectomies were grouped as the first (pancreatoduodenectomy-1) or second (pancreatoduodenectomy-2) pancreatoduodenectomy on a single day (ie, paired pancreatoduodenectomies) and as pancreatoduodenectomy-3 whenever 1 pancreatoduodenectomy was performed per day (ie, unpaired). Patients undergoing minimally invasive procedures were excluded. The primary outcomes were major morbidity (ie, Clavien-Dindo grade ≥IIIa) and mortality.
Results
Among 689 patients, 151 patients had undergone minimally invasive pancreatoduodenectomy, leaving 538 patients after open pancreatoduodenectomy for inclusion. The overall rate of major morbidity was 37.4% (n = 200/538) and in-hospital/30-day mortality 1.7% (n = 9/538). Overall, 136 (25.3%) patients were operated in 68 pancreatoduodenectomy-1/ pancreatoduodenectomy-2 pairs and 402 (74.7%) patients as unpaired pancreatoduodenectomy (pancreatoduodenectomy-3). No differences were found between pancreatoduodenectomy-1 and pancreatoduodenectomy-2 regarding the rates of major morbidity (35.3% vs 26.5%; P = .265) and mortality (1.5% vs 0%; P = .999). Between the 68 pancreatoduodenectomy-1/ pancreatoduodenectomy-2 pairs and the 402 unpaired pancreatoduodenectomies, the rates of major morbidity (30.9% vs 39.6%; P = .071) and mortality (0.7% vs 2.0%; P = .461) did not differ significantly. In multivariable logistic regression analysis, pancreatoduodenectomy-1 was not associated with major morbidity (odds ratio = 0.913 [95% confidence interval 0.515–1.620]; P = .756), whereas pancreatoduodenectomy-2 was associated with less major morbidity (odds ratio = 0.522 [95% confidence interval 0.277–0.983]; P = .045).
Conclusion
In a high-volume setting, performing 2 consecutive open pancreatoduodenectomies on a single operating day appears to be safe. This approach may be an option when logistically required.
Introduction
Pancreatoduodenectomy (PD) is among the most complex surgical operations. In patients undergoing PD, hospital volume is strongly and inversely related to postoperative mortality.
1
This has led to the concentration of PD to high-volume centers, which reduced mortality rates by nearly half in the Netherlands.2
,3
Nevertheless, PD remains associated with high complication rates up to 65.3%.4
As a result of the continuing centralization, the demand for pancreatic surgery in high-volume centers is growing. The latter is illustrated by the fact that resection rates for pancreatic cancer have doubled from 8.3% to 16.6% within a decade in the Netherlands.5
Additionally, the use of surgery in patients with initially unresectable pancreatic cancer after preoperative chemotherapy is increasing.6
, 7
, 8
However, this increase may be offset by the increased use of neoadjuvant therapy, which may decrease the resection rate in patients with borderline resectable pancreatic cancer.9
,10
Considering the complex nature of pancreatoduodenectomy and the management of the associated complications,11
,12
pancreatic surgery requires considerable employment of health care system resources.13
Therefore, one of the major concerns is whether high-volume institutions will remain capable of providing timely access to surgery, which may be particularly relevant for patients with malignant disease, characterized by its exceptionally aggressive biological behavior.14
It is known that efficient surgical treatment planning increases cost-effectiveness, reduces length of hospital stay and complication rates, and maximizes the use of (limited) operating resources.
15
As such, often driven by logistical restraints, several high-volume pancreatic centers have started programs in which 2 consecutive PDs are performed by the same surgical team on a single day. This practice may fulfill the requirement for prompt access to surgical care and prevents (unnecessary) waiting times for operation. However, data regarding the potential impact of this approach on surgical outcome and patient safety are scarce.16
,17
Therefore, this study aims to investigate the safety of 2 consecutive open PDs performed by the same surgical team during a single day.Methods
This study was performed in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology guidelines.
18
Study design
A retrospective analysis including all consecutive PDs in a single high-volume center (Amsterdam UMC, Amsterdam, the Netherlands) was performed. Data were obtained via the mandatory Dutch Pancreatic Cancer Audit,
19
as part of the Dutch Pancreatic Cancer Group.20
All patients who underwent a PD for any indication from January 2014 until December 2021 were identified. This included PDs performed at location Academic Medical Center of the Amsterdam University Medical Center (Jan 2014 to Jun 2021) and, thereafter, after the merger of both Hepato-Pancreato-Biliary teams of Amsterdam UMC, in location VUMC (final 6 months of the study period).Additional data were collected from local electronic medical records. Minimally invasive PDs (ie, robotic or laparoscopic), with or without conversion to an open approach, were excluded to prevent selection bias. Indications for minimally invasive surgery remained similar during the entire study period: patients with a body mass index <35 kg/m2, without a history of chronic or recurrent acute pancreatitis, and without signs of vascular involvement on a pancreas-dedicated computed tomography scan of maximally 4 weeks old. Hereby, these patients have a higher risk of postoperative pancreatic fistula (eg, small periampullary cancers, cysts) giving rise to selection bias.
Since 2014, a program existed in which 2 consecutive PDs were performed on 1 day by a single surgical team, consisting of 1 surgeon and a fellow or senior surgical resident, without rotation. Depending on their experience and the operation’s complexity, part of the operation could be performed by either the fellow or senior resident under direct supervision (eg, cholecystectomy, hepatico-jejunostomy, duodeno-jejunostomy), with the same surgeon present for the vast majority of both procedures. For both PDs, consistent availability of nursing and anesthesia teams was present, whereby scrub nurses were able to rotate out. At no time were operations performed without continuous and stable access to nursing and anesthesia teams. This operating day started at 8 AM with the surgical time out and was scheduled to end at approximately 6:30 PM. Typically, both PDs were finished within this time frame; although, in approximately 15% of PDs, the second procedure was finished later. Surgeons performing 2 consecutive PDs on a single operating day minimized their on-call responsibilities the day prior.
The PDs were listed as either the first (PD1) or second (PD2) PD on a single day (ie, paired PDs) or as PD3 whenever a single PD on a day (unpaired PDs) was performed. No specific selection criteria for allocation to either treatment group were maintained, except for patients undergoing minimally invasive (laparoscopic, robotic) PD, which were always scheduled as a single PD per day. All planned pairs are described, as there was no day where the surgical team did not proceed with the second procedure. Typically, unpaired PDs (PD3) were combined with a smaller surgical intervention (eg, laparoscopic distal pancreatectomy or laparoscopic cholecystectomy), whereby the PD was always planned first. Patients in the PD1 group were compared to those in the PD2 group. Additionally, all PD pairs (PD1 and PD2) were compared to the unpaired PDs (PD3).
Outcome measurements and definitions
The primary endpoints were major morbidity, defined as Clavien-Dindo Grade IIIa or higher,
21
and mortality, either in-hospital or, in case of earlier discharge, within 30 days after index surgery. The secondary endpoints included pancreatic surgery–specific complications (ie, postoperative pancreatic fistula, delayed gastric emptying, postpancreatectomy hemorrhage, bile leakage, and chyle leakage), all grade B/C as defined by the International Study Group for Pancreatic Surgery (ISGPS) and International Study Group of Liver Surgery.22
, 23
, 24
, 25
, 26
Complications were registered when they occurred during hospital stay or within 30 days after index surgery.All diagnosed conditions were classified in accordance with the World Health Organization definitions.
27
,28
The preoperative physical condition of patients was classified according to the American Society of Anesthesiologists–Physical Status.29
Pancreatoduodenectomy was defined following the ISGPS definition and comprised the “classical” Whipple procedure (ie, PD), pylorus-preserving pancreatoduodenectomy, and pylorus-resecting pancreatoduodenectomy.American Society of Anesthesiologists Physical Status Classification System; 2020.
https://www.asahq.org/standards-and-guidelines/asa-physical-status-classification-system
Date accessed: September 5, 2022
30
Portomesenteric vein resections were classified in accordance with the ISGPS guideline.31
Radical resection was defined following the Royal College of Pathologists definition as absence of tumor within 1 mm of any of resection and/or dissection margin.32
Macroscopic irradicality was intraoperatively assessed by the surgeon. Readmission was defined as admission to the hospital for any reason within 30 days after discharge. Patients who died in the hospital were excluded from the calculations regarding the readmission rate.Statistical analysis
Data analyses were performed with RStudio: Integrated Development Environment for R (software version 1.3.1093). Descriptive statistics were compiled to summarize patient characteristics. Pearson’s χ2 anlysis, or Fisher exact test when appropriate, was used to compare the categorical variables; the Mann-Whitney U test was used to compare the numerical variables. The categorical variables are presented as percentages and frequencies, and the numerical variables are presented as medians with corresponding IQRs.
Logistic regression analysis was performed to identify predictors for major morbidity. Multiple imputation, used to account for missing data, was based on 10 imputation sets and predictive mean matching using the mice package. Variations between observed and imputed data were visually assessed using density plots. The results were pooled using Rubin’s rule.
34
The results of the logistic regression analysis are presented in odds ratios (ORs) with the corresponding 95% CIs. In multivariable regression models, a backward stepwise selection was used for removal of variables. Known prognostic factors and factors with a P value < .200 on univariable analysis were included in backward stepwise selection in the multivariable analysis. The included variables with the highest nonsignificant P value were stepwise eliminated, until all remaining variables were statistically significant.- Heymans M.W.
- Eekhout I.
Applied missing data analysis with SPSS and (R)Studio; 2019.
https://bookdown.org/mwheymans/bookmi/
Date accessed: September 5, 2022
35
Results
In total, 689 patients underwent PD during the study period. After excluding 151 patients (21.9%) undergoing minimally invasive PD, with or without conversion to an open approach, the final study cohort comprised 538 patients after open PD. This included 136 PDs (n = 136/538, 25.3%) in 68 pairs, assigned to either the PD1 (n = 68) or PD2 (n = 68) group, and 402 patients (74.7%) undergoing an “unpaired” PD (PD3).
Baseline characteristics
Of all 538 included PDs, the median age at time of surgery was 68 years (IQR 58.0–74.0), and 64 patients (11.9%) had an Eastern Cooperative Oncology Group performance status of ≥2. Overall, 490 patients (91.1%) were diagnosed with malignant disease, according to final histopathology, of which the majority concerned pancreatic cancer (n = 274/490, 55.9%). Preoperative chemotherapy and chemoradiotherapy was administered in 112 patients (20.8%). Table I includes all baseline demographics. No differences were found in the baseline characteristics between PD1 and PD2. Most of the paired PDs (n = 64/68, 94.1%) were performed by, or supervised by, the same senior surgeon (O.R.B.), whom was within his fifth decade of life when performing these operations, having 20 years of experience. The baseline characteristics of paired PDs did not differ significantly from patients in the PD3 group, except for the type of PD performed. When assessing the baseline characteristics for patients with cancer specifically, no differences were observed within the paired PDs or between all paired PDs and PD3 groups. Supplementary Table S1 includes the baseline characteristics specifically for all patients with cancer.
Table IClinicopathological characteristics
| Baseline characteristics | Overall (n = 538) | Total cohort | |||||
|---|---|---|---|---|---|---|---|
| PD1 (n = 68) | PD2 (n = 68) | P value | Paired PDs (n = 138) | PD3 (n = 402) | P value | ||
| Sex, male, n (%) | 277 (51.5) | 37 (54.4) | 41 (60.3) | .488 | 78 (57.4) | 199 (49.5) | .113 |
| Age, y, median (IQR) | 68.0 (58.0–74.0) | 67.5 (55.0–74.2) | 67.5 (61.8–72.2) | .504c | 67.5 (57.8–73.2) | 68.0 (58.0–74.0) | .780 |
| ECOG PS ≥2, n (%) | 64 (11.9) | 9 (13.2) | 10 (14.7) | .805 | 19 (14) | 45 (11.2) | .393 |
| BMI, kg/m 2 , median (IQR) | 24.5 (22.3–27.4) | 24.6 (22.4–28.0) | 25.3 (22.6–29.4) | .352c | 25.2 (22.4–28.4) | 24.4 (22.3–26.9) | .074 |
| ASA score (>2), n (%) | 133 (25.1) | 12 (17.6) | 17 (25.0) | .295 | 29 (21.3) | 104 (26.4) | .239 |
| Perioperative characteristics | |||||||
| Preoperative therapy, n (%) | 112 (20.8) | 11 (16.2) | 10 (14.7) | .904 | 21 (15.4) | 91 (22.6) | .124 |
| Chemotherapy | 67 (12.5) | 6 (8.8) | 4 (5.8) | - | 10 (7.4) | 57 (14.2) | - |
| Chemoradiotherapy | 40 (7.4) | 5 (7.3) | 6 (8.8) | - | 11 (8.1) | 29 (7.2) | - |
| Radiotherapy | 4 (0.7) | 0 (0.0) | 0 (0.0) | - | 0 (0.0) | 4 (1.0) | - |
| Pancreatoduodenectomy, n (%) | - | - | - | .063 | - | - | .005 |
| Pylorus preserving | 397 (73.8) | 62 (91.2) | 52 (76.5) | - | 114 (83.8) | 283 (70.4) | - |
| Pylorus resecting | 61 (11.3) | 2 (2.9) | 5 (7.4) | - | 7 (5.1) | 54 (13.4) | - |
| Classical Whipple | 80 (14.9) | 4 (5.9) | 11 (16.2) | - | 15 (11.0) | 65 (16.2) | - |
| Pancreatico-enterostomosis, n (%) | - | - | - | .619 | - | - | .081 |
| Pancreatico-jejunostomy | 509 (97.7) | 68 (100.0) | 67 (98.5) | - | 135 (99.3) | 374 (93.0) | - |
| Other/unknown | 12 (2.3) | 0 (0.0) | 0 (0.0) | - | 0 (0.0) | 12 (3.0) | - |
| Colectomy, n (%) | 7 (1.3) | 1 (1.5) | 1 (1.5) | >.999 | 2 (1.5) | 5 (1.2) | >.999 |
| (Sub)total gastrectomy, n (%) | 5 (0.9) | 1 (1.5) | 1 (1.5) | >.999 | 2 (1.5) | 3 (0.7) | .605 |
| Portomesenteric venous resection, n (%) | 123 (22.9) | 10 (14.7) | 18 (26.5) | .090 | 28 (20.6) | 95 (23.7) | .457 |
| Type 1–2 | 86 (16.0) | 9 (13.2) | 12 (17.6) | - | 21 (15.4) | 65 (16.2) | - |
| Type 3–4 | 37 (6.9) | 1 (1.5) | 6 (8.8) | - | 7 (5.1) | 30 (7.5) | - |
| Arterial resection, n (%) | 11 (2.0) | 1 (1.5) | 1 (1.5) | >.999 | 2 (1.5) | 9 (2.2) | .738 |
| Aspect of the pancreas, n (%) | - | - | - | .672 | - | - | .887 |
| Normal/soft | 255 (48.1) | 33 (48.5) | 34 (50.7) | - | 67 (49.6) | 188 (47.6) | - |
| Hard/fibrotic | 240 (45.3) | 32 (47.1) | 28 (41.8) | - | 60 (44.4) | 180 (45.6) | - |
| Unknown | 35 (6.6) | 3 (4.4) | 5 (7.5) | - | 8 (5.9) | 27 (6.8) | - |
| Diameter pancreatic duct, mm, median (IQR) | 3.0 (2.0–4.0) | 3.0 (2.2–4.0) | 3.0 (2.0–4.0) | .635 | 3.0 (2.0–4.0) | 3.0 (2.0–4.0) | .155 |
| Pathology | |||||||
| Diagnoses, n (%) | - | - | - | .335 | - | - | .386 |
| Adenocarcinoma | 420 (78.1) | 52 (76.5) | 47 (69.1) | - | 99 (72.8) | 321 (79.9) | - |
| Noninvasive IPMN | 39 (7.2) | 7 (10.3) | 5 (7.5) | - | 12 (8.8) | 27 (6.7) | - |
| PNET | 21 (3.9) | 2 (2.9) | 3 (4.4) | - | 5 (3.7) | 16 (4.0) | - |
| Chronic pancreatitis | 12 (2.2) | 0 (0.0) | 4 (5.9) | - | 4 (2.9) | 8 (2.0) | - |
| Other | 43 (8.0) | 6 (8.8) | 9 (13.2) | - | 15 (11.0) | 28 (7.0) | - |
ASA, American Society of Anesthesiology score; BMI, body mass index; ECOG PS, Eastern Cooperative Oncology Group performance status; IPMN, intraductal papillary mucinous neoplasm; N, number of patients; PD, pancreatoduodenectomy; PNET, pancreatic neuroendocrine tumor.
∗ Missing data: BMI (n = 34), ECOG (n = 1), origin (n = 6), ASA (n = 8), preoperative therapy (n = 16), pancreatic enterostomosis (n = 17), aspect of the pancreas (n = 8), diameter pancreatic duct (n = 162), diagnoses (n = 3).
† P value comparison PD1 versus PD2.
‡ P value comparison paired PDs (PD1+PD2) versus unpaired PDs (PD3).
§ Pearson’s χ2 test
| Fisher exact test
¶ Mann-Whitney U test
# Assessed intraoperatively by the operating surgeon.
Surgical outcome–total cohort
Overall, the rate of major morbidity was 37.4% (n = 200/538) and mortality 1.7% (n = 9/538). The postoperative outcome measures of the complete study cohort are summarized in Supplementary Table S2.
Surgical outcome–paired PDs
No differences were observed between PD1 and PD2 for median operation time (231 minutes [IQR 192–263] vs 224 minutes [IQR 200–274]; P = .718) and intraoperative blood loss (350 mL [IQR 250–640] vs 387.5 mL [IQR 250–763]; P = .426). Additionally, major morbidity (35.3% vs 26.5%; P = .265) and mortality (1.5% [n = 1/68] vs 0%; P = .999) were not different between PD1 and PD2, respectively. The postoperative outcome measures of all paired PDs are found in Table II.
Table IIComparison of postoperative outcome measures between paired PDs (PD1 and PD2)
| Outcome measures | PD1 (n = 68) | PD2 (n = 68) | P value |
|---|---|---|---|
| Operation time, min, median [IQR] | 230.5 (192.0–263.3) | 224.0 (200.0–274.0) | .718 |
| Blood loss, mL, median (IQR) | 350.0 (250.0–640.0) | 387.5 (250.0–762.5) | .426 |
| POPF (grade B/C), n (%) | 10 (14.9) | 10 (14.7) | .971 |
| Chyle leakage (grade B/C), n (%) | 4 (7.4) | 1 (1.8) | .198 |
| PPH (grade B/C), n (%) | 8 (11.9) | 5 (7.4) | .366 |
| Bile leakage (grade B/C), n (%) | 6 (9.0) | 5 (7.4) | .841 |
| DGE (grade B/C), n (%) | 10 (14.9) | 11 (16.2) | .734 |
| Major morbidity, n (%) | 24 (35.3) | 18 (26.5) | .265 |
| Relaparotomy | 4 (5.9) | 2 (2.9) | .679 |
| Organ failure | 5 (7.4) | 2 (2.9) | .441 |
| MCU/ICU admission | 11 (16.2) | 3 (4.4) | .024 |
| Mortality, n (%) | 1 (1.5) | 0 (0.0) | .999 |
| Hospital stay, d, median (IQR) | 8.0 (7.0–14.0) | 9.0 (7.0–14.0) | .697 |
| Readmission, n (%) | 14 (20.9) | 11 (16.2) | .480 |
Percentages may not add up due to rounding and missing data.
DGE, delayed gastric emptying; ICU intensive care unit; MCU, medium-care unit; N, number of patients; PD, pancreatoduodenectomy; POPF, postoperative pancreatic fistula; PPH, postpancreatectomy hemorrhage.
∗ Missing data: operation time (n = 41), blood loss (n = 11), POPF (n = 1), chyle leakage (n = 14), PPH (n = 1), bile leakage (n = 1), DGE (n = 1), hospital stay (n = 1), readmission (n = 1).
† Pearson’s χ2
‡ Fisher exact test
§ Mann-Whitney U test
Surgical outcome–paired versus unpaired PD
Compared to all 136 paired PDs, operating time in the unpaired PDs (PD3) was longer (228 min [IQR 194–272] vs 281 min [IQR 242–358]; P < .001) with increased blood loss (350 mL [IQR 250–700] vs 500 mL [IQR 300–900]; P = .013). No significant differences were seen for major morbidity (30.9% [n = 42/136] vs 39.6% [n = 159/402]; P = .071) and mortality (0.7% [n = 1/136] vs 2.0% [n = 8/402]; P = .461]). Table III compares the postoperative outcomes for the paired and unpaired PDs.
Table IIIComparison of postoperative outcome measures between paired PDs and unpaired PDs
| Outcome measures | Paired PDs (PD1, PD2) (n = 136) | Unpaired PDs (PD3) (n = 402) | P value |
|---|---|---|---|
| Operation time, min, median (IQR) | 228.0 (193.5–272.0) | 281.5 (242.0–358.0) | < .001 |
| Blood loss, mL, median (IQR) | 350.0 (250.0–700.0) | 500.0 (300.0–900.0) | .013 |
| POPF (grade B/C), n (%) | 20 (14.8) | 73 (18.2) | .362 |
| Chyle leakage (grade B/C), n (%) | 5 (4.5) | 15 (4.1) | .804 |
| PPH (grade B/C), n (%) | 13 (9.6) | 36 (9.0) | .826 |
| Bile leakage (grade B/C), n (%) | 11 (8.1) | 34 (8.5) | .893 |
| DGE (grade B/C), n (%) | 21 (15.6) | 85 (21.2) | .151 |
| Major morbidity, n (%) | 42 (30.9) | 159 (39.6) | .071 |
| Relaparotomy | 6 (4.4) | 26 (6.5) | .735 |
| Organ failure | 7 (5.1) | 24 (6.0) | .726 |
| MCU/ICU admission | 14 (10.3) | 42 (10.4) | .932 |
| Mortality, n (%) | 1 (0.7) | 8 (2.0) | 461 |
| Hospital stay, d, median (IQR) | 9.0 (7.0–14.0) | 10.0 (7.0–17.0) | .566 |
| Readmission, n (%) | 25 (18.5) | 64 (16.2) | .144 |
Percentages may not add up due to rounding and missing data.
DGE, delayed gastric emptying; ICU, intensive care unit; MCU, medium-care unit; N, number of patients; PD, pancreatoduodenectomy; POPF, postoperative pancreatic fistula; PPH, postpancreatectomy hemorrhage.
∗ Missing data: operation time (n = 133), blood loss (n = 28), POPF (n = 3), chyle leakage (n = 133), postpancreatectomy hemorrhage (n = 3), bile leakage (n = 4), DGE (n = 3), hospital stay (n = 15), readmission (n = 12).
† Pearson’s χ2
‡ Fisher exact test
§ Mann-Whitney U test
Predictors for major morbidity
In multivariable logistic regression analysis, PD1 was not associated with major morbidity (OR = 0.913 [95% CI 0.515–1.620]; P = .756), whereas PD2 (OR = 0.522 [95% CI 0.277–0.983]; P = .045) was protective for the occurrence of major morbidity. Furthermore, female sex, age >70 years, body mass index >25.0 kg/m2, duration of surgery, and malignant disease were identified as independent prognostic factors for major morbidity. The complete univariable and multivariable logistic regression analysis are covered in Table IV.
Table IVLogistic regression analysis–major morbidity
| Univariable analysis | Multivariable analysis | |||||
|---|---|---|---|---|---|---|
| Variables | OR | 95% CI | P value | OR | 95% CI | P value |
| Female sex | ||||||
| Male | 1 [reference] | - | - | 1 [reference] | - | - |
| Female | 0.649 | 0.456–0.923 | .017 | 0.656 | 0.449–0.960 | .030 |
| Age, y | ||||||
| <70 | 1 [reference] | - | - | 1 [reference] | - | - |
| ≥70 | 1.634 | 1.147 - 2.329 | .007 | 1.788 | 1.219–2.624 | .003 |
| BMI, kg/m2 | ||||||
| 18.5–25.0 | 1 [reference] | - | - | 1 [reference] | - | - |
| <18.5 | 2.011 | 0.731–5.534 | 0.177 | 2.320 | 0.786–6.849 | .128 |
| >25.0 | 1.702 | 1.176–2.463 | .005 | 1.704 | 1.160–2.505 | .007 |
| ECOG PS | ||||||
| <2 | 1 [reference] | - | - | - | - | - |
| ≥2 | 0.861 | 0.498–1.490 | .593 | - | - | - |
| Preoperative therapy | ||||||
| No | 1 [reference] | - | - | - | - | - |
| Yes | 0.780 | 0.501–1.213 | .270 | - | - | - |
| PD3 | 1 [reference] | - | - | 1 [reference] | - | - |
| PD1 | 0.834 | 0.488–1.425 | .506 | 0.913 | 0.515–1.620 | .756 |
| PD2 | 0.550 | 0.303–0.961 | .042 | 0.522 | 0.277–0.983 | .045 |
| Vascular resection | ||||||
| No | 1 [reference] | - | - | - | - | - |
| Yes | 0.641 | 0.416–0.980 | .040 | - | - | - |
| Operation time | 1.002 | 1.000–1.004 | .043 | 1.002 | 1.000–1.005 | .061 |
| Malignant disease | ||||||
| No | 1 [reference] | - | - | 1 [reference] | - | - |
| Yes | 0.523 | 0.272–1.007 | .053 | 0.360 | 0.175–0.743 | .006 |
BMI, body mass index; ECOG PS, Eastern Cooperative Oncology Group performance status; OR, odds ratio; PD, pancreatoduodenectomy.
∗ Imputed data: BMI (n = 34), ECOG PS (n = 1), preoperative therapy (n = 16), operation time (n = 133), malignant disease (n = 9).
Discussion
This retrospective analysis in a high-volume center suggests that performing 2 open PDs on a single (extended) operating day by the same surgical team is safe in terms of major morbidity and mortality. The outcomes did not differ significantly both within the PD pairs (PD1 and PD2) and between the paired (PD1–2) and unpaired (PD3) PDs. Also, when corrected for confounders, performing 2 consecutive PDs was not associated with worse outcome compared to performing a single PD.
Two retrospective, single-center series have previously investigated the effect of performing 2 PDs on a single day to optimize operation room use as a potential solution to the surgical waiting list prolongation.
16
,17
First, in a propensity score matched analysis, a single-center study from Indianapolis revealed no impact of performing 2 consecutive open PDs on safety and quality outcome in a cohort of 661 patients. However, the number of paired PDs (n = 25) performed in the 5-year inclusion period was relatively low, precluding (multivariable) logistic regression analysis.17
Second, a retrospective single-center study from Taiwan assessed 101 paired PDs within a cohort of 1,068 patients and found this practice to be safe as well.16
Unfortunately, the latter study only compared the second PD of a single surgical day with a combination of the first PD and “standalone” PDs (ie, PD2 vs PD1+PD3). Therefore, this study did not answer the questions whether the outcomes of the second PD are different from the first PD and whether the outcomes differ from standalone (ie, PD3) PD. This hampers the translation of these results into clinical practice.Performing 2 PDs on a single day may be seen as a daunting task leading to fatigue and work stress for those involved. Wu et al used the surgery task load index questionnaire (a recognized technique in clinical research to assess the surgery-related workload) to quantify and compare surgical workload between the first and second PD performed by the same surgeon on a single day.
36
Although the perceived surgical workload was increased in the subscales of physical and temporal demand, a relationship between an increased workload and poor surgical outcome was not established.16
On the other hand, a retrospective analysis from China revealed that performing overtime pancreatoduodenectomy, defined as start of surgery between 5 PM and 10 PM, increases the incidence of pancreatic fistula (32.8% vs 15.8%; P = .005).37
When combined with the present study, these findings confirm that performing 2 PDs on a single day is a safe and feasible possible option for high-volume centers coping with logistical challenges. Especially in the light of growing surgical demands on single institutions, known for its challenge to maintain acceptable waiting times, optimization of surgical procedure planning may offer timely access to highly complex care.38
, 39
, 40
Nevertheless, this planning should be done carefully, particularly with regard to the increased use of surgery in patients with locally advanced pancreatic cancer.41
,42
In case of such complex resections,43
it seems prudent to plan such procedures as a single PD per day.The results of this study should be interpreted in the light of several limitations. First, most of the paired PDs were performed by a single experienced surgeon, or under his supervision, hampering external validity. Although 2 procedures could be performed safely within a single extended surgical program, this should never be seen as a goal in itself. Therefore, the authors do not advocate to perform 2 Whipple procedures per day as a standard, but rather as a possible solution to optimize the use of operation facilities in times of increasing pressure on patient waiting lists and medical care resources. Second, although no formal patient selection process existed, the appointment of patients to either treatment group was per definition influenced by the exclusion of minimally invasive procedures. As minimally invasive PDs were performed in patients without vascular contact during essentially the entire study period, all remaining open procedures were relatively more complex (with a need for portomesenteric venous resection in 22.9% of the total cohort). Additionally, the results may be influenced by treatment allocation bias as the most complex procedures are likely to be scheduled in the PD3 group to prevent time constraints. By including surrogate markers for the complexity of surgery in the logistic regression analysis (eg, operation time, presence of malignant disease, and necessity to perform vascular resection), we sought to overcome this bias. Still, the second PD of the combination day was associated with a trend toward more vascular resections compared to the first PD of the day, suggesting the presence of patient selection–related confounders introduced by surgical planning. However, this did not lead to increased operative times as these were similar for the first and second procedure. Third, this study was conducted in a single high-volume center as reflected by the low (1.7%) mortality rate. Therefore, these results may be generalizable only to other high-volume institutions.
In conclusion, because of the continuing centralization in combination with the rising need for pancreatic (cancer) surgery, more PDs are performed at high-volume centers. Performing 2 consecutive PDs on a single working day by the same surgical team was not associated with increased major morbidity or mortality compared to performing a single PD per day. This approach could offer a solution for high-volume centers coping with logistical challenges due to the rising demand for time-consuming and complex surgical procedures.
Funding/Support
This research did not receive any specific funding from any agencies in the public, commercial, or not-for-profit areas.
Conflict of interest/Disclosure
The authors have no conflicts of interests or disclosures to report.
Acknowledgments
The authors would like to thank Susan van Dieren (clinical statistician, Department of Statistics and Epidemiology, Amsterdam UMC, University of Amsterdam) for her contribution to construct and support the statistical analysis.
Supplementary Materials
- Supplementary Table 1
References
- Effect of hospital volume on surgical outcomes after pancreaticoduodenectomy: a systematic review and meta-analysis.Ann Surg. 2016; 263: 664-672
- Impact of centralization of pancreatic cancer surgery on resection rates and survival.Br J Surg. 2014; 101: 1000-1005
- Impact of nationwide centralization of pancreaticoduodenectomy on hospital mortality.Br J Surg. 2012; 99: 404-410
- Benchmarks in pancreatic surgery: a novel tool for unbiased outcome comparisons.Ann Surg. 2019; 270: 211-218
- Effect of centralization and regionalization of pancreatic surgery on resection rates and survival.Br J Surg. 2021; 108: 826-833
- Cancer statistics, 2022.CA Cancer J Clin. 2022; 72: 7-33
- Trends in the utilization of neoadjuvant therapy for pancreatic ductal adenocarcinoma.J Surg Oncol. 2021; 123: 1432-1440
- Neoadjuvant FOLFIRINOX in patients with borderline resectable pancreatic cancer: a systematic review and patient-level meta-analysis.J Natl Cancer Inst. 2019; 111: 782-794
- Neoadjuvant chemoradiotherapy versus upfront surgery for resectable and borderline resectable pancreatic cancer: long-term results of the Dutch Randomized PREOPANC Trial.J Clin Oncol. 2022; 40: 1220-1230
- Meta-analysis comparing upfront surgery with neoadjuvant treatment in patients with resectable or borderline resectable pancreatic cancer.Br J Surg. 2018; 105: 946-958
- Impact of complications after pancreatoduodenectomy on mortality, organ failure, hospital stay, and readmission: analysis of a nationwide audit.Ann Surg. 2022; 275: e222-e228
- Failure to rescue after pancreatoduodenectomy: a transatlantic analysis.Ann Surg. 2021; 274: 459-466
- Hospital costs of complications after a pancreatoduodenectomy.HPB (Oxford). 2015; 17: 723-731
- Time to progression of pancreatic ductal adenocarcinoma from low-to-high tumour stages.Gut. 2015; 64: 1783-1789
- Enhancing the cost-effectiveness of surgical care in pancreatic cancer: a systematic review and cost meta-analysis with trial sequential analysis.HPB (Oxford). 2022; 24: 309-321
- The effect of performing two pancreatoduodenectomies by a single surgical team in one day on surgeons and patient outcomes.HPB (Oxford). 2020; 22: 1185-1190
- Two open Whipples a day: excessive or efficient.Am J Surg. 2020; 219: 390-393
- The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies.J Clin Epidemiol. 2008; 61: 344-349
- Nationwide prospective audit of pancreatic surgery: design, accuracy, and outcomes of the Dutch Pancreatic Cancer Audit.HPB (Oxford). 2017; 19: 919-926
- Establishing and coordinating a nationwide multidisciplinary study group: lessons learned by the Dutch Pancreatic Cancer Group.Ann Surg. 2020; 271: e102-e104
- The Clavien-Dindo classification of surgical complications: five-year experience.Ann Surg. 2009; 250: 187-196
- The 2016 update of the International Study Group (ISGPS) definition and grading of postoperative pancreatic fistula: 11 years after.Surgery. 2017; 161: 584-591
- Delayed gastric emptying (DGE) after pancreatic surgery: a suggested definition by the International Study Group of Pancreatic Surgery (ISGPS).Surgery. 2007; 142: 761-768
- Postpancreatectomy hemorrhage (PPH): an International Study Group of Pancreatic Surgery (ISGPS) definition.Surgery. 2007; 142: 20-25
- Definition and classification of chyle leak after pancreatic operation: a consensus statement by the International Study Group on Pancreatic Surgery.Surgery. 2017; 161: 365-372
- Bile leakage after hepatobiliary and pancreatic surgery: a definition and grading of severity by the International Study Group of Liver Surgery.Surgery. 2011; 149: 680-688
- International consensus guidelines 2012 for the management of IPMN and MCN of the pancreas.Pancreatology. 2012; 12: 183-197
- The 2019 WHO classification of tumours of the digestive system.Histopathology. 2020; 76: 182-188
- American Society of Anesthesiologists Physical Status Classification System; 2020.https://www.asahq.org/standards-and-guidelines/asa-physical-status-classification-systemDate accessed: September 5, 2022
- Extended pancreatectomy in pancreatic ductal adenocarcinoma: definition and consensus of the International Study Group for Pancreatic Surgery (ISGPS).Surgery. 2014; 156: 1-14
- Borderline resectable pancreatic cancer: a consensus statement by the International Study Group of Pancreatic Surgery (ISGPS).Surgery. 2014; 155: 977-988
- Dataset for the histopathological reporting of carcinomas of the pancreas, ampulla of Vater and common bile duct.The Royal College of Pathologists, 2010
- Integrated Development for R. RStudio, PBC, Boston, MA.(Accessed 5 Sept)
- Applied missing data analysis with SPSS and (R)Studio; 2019.https://bookdown.org/mwheymans/bookmi/Date accessed: September 5, 2022
- Understanding logistic regression analysis.Biochem Med (Zagreb). 2014; 24: 12-18
- Development and validation of a surgical workload measure: the surgery task load index (SURG-TLX).World J Surg. 2011; 35: 1961-1969
- Effect of overtime pancreaticoduodenectomy on the short-term prognosis of patients.World J Gastrointest Surg. 2022; 14: 419-428
- Optimization and planning of operating theatre activities: an original definition of pathways and process modeling.BMC Med Inform Decis Mak. 2015; 15: 38
- Surgical scheduling: a lean approach to process improvement.Aorn j. 2014; 99: 147-159
- Optimization of surgery sequencing and scheduling decisions under uncertainty.Health Care Manag Sci. 2007; 10: 13-24
- Locally advanced pancreatic cancer: work-up, staging, and local intervention strategies.Cancers (Basel). 2019; 11: 976
- Meta-analysis of FOLFIRINOX-based neoadjuvant therapy for locally advanced pancreatic cancer.Medicine (Baltimore). 2021; 100e24068
- Anatomic criteria determine resectability in locally advanced pancreatic cancer.Ann Surg Oncol. 2022; 29: 401-414
Article info
Publication history
Published online: February 25, 2023
Accepted:
January 17,
2023
Publication stage
In Press Corrected ProofIdentification
Copyright
© 2023 The Author(s). Published by Elsevier Inc.
User license
Creative Commons Attribution (CC BY 4.0) | How you can reuse 
Elsevier's open access license policy
Creative Commons Attribution (CC BY 4.0)
Permitted
- Read, print & download
- Redistribute or republish the final article
- Text & data mine
- Translate the article
- Reuse portions or extracts from the article in other works
- Sell or re-use for commercial purposes
Elsevier's open access license policy
