Martijn Willem Jan Stommel
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Martijn Willem Jan Stommel has a total of 731 co-authors. Here are the top co-authors by number of shared publications:
- Marc Gerard Besselink
Department of Surgery, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
28 shared publications - Richard Peter Gerardus Ten Broek
Department of Surgery, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands
19 shared publications - Olivier Rc Busch
Department of Surgery, Amsterdam University Medical Center, Amsterdam, the Netherlands
17 shared publications - Vincent E. de Meijer
Section of Hepatobiliary Surgery and Liver Transplantation, Department of Surgery, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands
15 shared publications - Judith de Vos-Geelen
Div. Medical Oncology, Dept. Internal Medicine, GROW - Research Institute for Oncology & Reproduction, Maastricht University Medical Centre, Maastricht, the Netherlands
14 shared publications - Bas Groot Koerkamp
Department of Surgery, Erasmus Medical Center, Rotterdam, the Netherlands
12 shared publications
Latest Publications
- Open Access
- 9 October 2026
- Published by Elsevier in European Journal of Cancer
- Vol. 248, 117776
Survival benefit of recurrence-focused treatment after surgery for pancreatic ductal adenocarcinoma: A propensity score matched analysis of patients from a Dutch National and US institutional cohort
- 9 October 2026
- Published by Elsevier in European Journal of Cancer
- Vol. 248, 117776
Open AccessSurvival benefit of recurrence-focused treatment after surgery for pancreatic ductal adenocarcinoma: A propensity score matched analysis of patients from a Dutch National and US institutional cohort
- Open Access
- 7 October 2026
- Published by Wolters Kluwer Health in Annals of Surgery
- Vol. 284 (5) , 993-1001
Pan-European Training Program for Robotic Pancreatoduodenectomy (LEARNBOT)
Objective: To assess the feasibility of and clinical outcomes after a pan-European training program for robotic pancreatoduodenectomy (RPD). Summary Background Data: The implementation of RPD into clinical practice comes with considerable challenges, leading to concerns about patient safety, oncological outcomes, and costs. To address this, a structured training program was implemented in which 2 surgeons at each of the 20 sites were trained and maintained as a consistent operative team. Study Design: A structured training program for RPD was implemented, and outcomes were prospectively collected in 20 European high-volume centers, all without previous RPD experience, from 12 countries. The program included a video library, biotissue simulation (pancreaticojejunostomy and hepaticojejunostomy), off-site RPD case observations, and on-site proctoring. Surgeons’ simulation anastomoses were assessed using Objective Structured Assessment of Technical Skills (OSATS) and technical errors. The primary endpoint was intra- and postoperative patient outcome. A key secondary analysis assessed safety using OSATS scores and error counts during the biotissue training.. Results: Overall, 486 RPD procedures were performed. Median intraoperative blood loss was 200 mL (IQR: 100–400), operative time was 475 minutes (IQR: 414–553.3), and the conversion rate was 18.3% (n=89). The rate of Clavien-Dindo grade ≥III complications was 31.4% (n=153), POPF was 21.2% (n=103), delayed gastric emptying was 15% (n=73), postpancreatectomy hemorrhage was 8.6% (n=42), and in-hospital/30-day mortality was 2.7% (n=13). Median length of stay was 12 days (IQR: 8–20). During simulation training, OSATS scores significantly improved (0.350 points/attempt ( ρ ), P P P =0.006]). Conclusions: We report on the first successfully completed international training program for RPD. Simulation established a standardized technical competency, although a learning-curve effect (POPF) remained detectable per center. These data highlight that safe adoption of complex procedures requires structured, team-based, proctored training beyond simulation alone.
- 7 October 2026
- Published by Wolters Kluwer Health in Annals of Surgery
- Vol. 284 (5) , 993-1001
Open AccessPan-European Training Program for Robotic Pancreatoduodenectomy (LEARNBOT)
Objective: To assess the feasibility of and clinical outcomes after a pan-European training program for robotic pancreatoduodenectomy (RPD). Summary Background Data: The implementation of RPD into clinical practice comes with considerable challenges, leading to concerns about patient safety, oncological outcomes, and costs. To address this, a structured training program was implemented in which 2 surgeons at each of the 20 sites were trained and maintained as a consistent operative team. Study Design: A structured training program for RPD was implemented, and outcomes were prospectively collected in 20 European high-volume centers, all without previous RPD experience, from 12 countries. The program included a video library, biotissue simulation (pancreaticojejunostomy and hepaticojejunostomy), off-site RPD case observations, and on-site proctoring. Surgeons’ simulation anastomoses were assessed using Objective Structured Assessment of Technical Skills (OSATS) and technical errors. The primary endpoint was intra- and postoperative patient outcome. A key secondary analysis assessed safety using OSATS scores and error counts during the biotissue training.. Results: Overall, 486 RPD procedures were performed. Median intraoperative blood loss was 200 mL (IQR: 100–400), operative time was 475 minutes (IQR: 414–553.3), and the conversion rate was 18.3% (n=89). The rate of Clavien-Dindo grade ≥III complications was 31.4% (n=153), POPF was 21.2% (n=103), delayed gastric emptying was 15% (n=73), postpancreatectomy hemorrhage was 8.6% (n=42), and in-hospital/30-day mortality was 2.7% (n=13). Median length of stay was 12 days (IQR: 8–20). During simulation training, OSATS scores significantly improved (0.350 points/attempt ( ρ ), P P P =0.006]). Conclusions: We report on the first successfully completed international training program for RPD. Simulation established a standardized technical competency, although a learning-curve effect (POPF) remained detectable per center. These data highlight that safe adoption of complex procedures requires structured, team-based, proctored training beyond simulation alone.
- Open Access
- 3 August 2026
- Published by Oxford University Press (OUP) in British Journal of Surgery
- Vol. 113 (8)
A paradox in the quality of care in centralization of pancreatic cancer surgery
- 3 August 2026
- Published by Oxford University Press (OUP) in British Journal of Surgery
- Vol. 113 (8)
Open AccessA paradox in the quality of care in centralization of pancreatic cancer surgery
- Open Access
- 3 July 2026
- Published by Oxford University Press (OUP) in BJS Open
- Vol. 10 (4)
Hospital costs associated with complications following pancreatoduodenectomy: systematic review
Background: Pancreatoduodenectomy is associated with postoperative complications in nearly half of all patients. Understanding the financial burden of these complications is relevant to supporting initiatives to improve patient outcome and resource allocation. A contemporary systematic review evaluating the costs of complications following pancreatoduodenectomy is lacking. Methods: A systematic review was conducted following PRISMA 2020 guidelines in PubMed, Embase, and Cochrane Central Register of Controlled Trials from inception to September 2024. Studies reporting on the hospital costs (primary outcome) of complications following pancreatoduodenectomy were included. Costs were adjusted for inflation and purchasing power parity to 2023 Dutch euros (€). Data extraction and risk-of-bias assessments were conducted independently by two reviewers. Results: Overall, 46 studies with 111 215 patients undergoing pancreatoduodenectomy from 13 countries were included. Postoperative pancreatic fistula grade B increased costs by €1998 to €25 228, and grade C by €21 268 to €112 824 versus no clinically relevant postoperative pancreatic fistula (mean or median values). Delayed gastric emptying grade B increased costs by €10 836 to €22 571 and grade C by €28 707 to €61 818 versus no clinically relevant delayed gastric emptying. Postpancreatectomy haemorrhage (various definitions) increased costs by €29 447 to €48 380. Biliary leak grade B increased costs by €7616 and grade C by €22 727. Clavien–Dindo complication grade III added €2785 to €20 294 compared with grade II, and grade IV added €25 190 to €63 644 compared with grade III. Conclusion: Complications following pancreatoduodenectomy increase costs substantially, exceeding €100 000 for a single postoperative pancreatic fistula grade C. These costs can now be used to design cost-effective prophylactic and therapeutic interventions aiming to improve patient outcomes together with optimal use of healthcare resources.
- 3 July 2026
- Published by Oxford University Press (OUP) in BJS Open
- Vol. 10 (4)
Open AccessHospital costs associated with complications following pancreatoduodenectomy: systematic review
Background: Pancreatoduodenectomy is associated with postoperative complications in nearly half of all patients. Understanding the financial burden of these complications is relevant to supporting initiatives to improve patient outcome and resource allocation. A contemporary systematic review evaluating the costs of complications following pancreatoduodenectomy is lacking. Methods: A systematic review was conducted following PRISMA 2020 guidelines in PubMed, Embase, and Cochrane Central Register of Controlled Trials from inception to September 2024. Studies reporting on the hospital costs (primary outcome) of complications following pancreatoduodenectomy were included. Costs were adjusted for inflation and purchasing power parity to 2023 Dutch euros (€). Data extraction and risk-of-bias assessments were conducted independently by two reviewers. Results: Overall, 46 studies with 111 215 patients undergoing pancreatoduodenectomy from 13 countries were included. Postoperative pancreatic fistula grade B increased costs by €1998 to €25 228, and grade C by €21 268 to €112 824 versus no clinically relevant postoperative pancreatic fistula (mean or median values). Delayed gastric emptying grade B increased costs by €10 836 to €22 571 and grade C by €28 707 to €61 818 versus no clinically relevant delayed gastric emptying. Postpancreatectomy haemorrhage (various definitions) increased costs by €29 447 to €48 380. Biliary leak grade B increased costs by €7616 and grade C by €22 727. Clavien–Dindo complication grade III added €2785 to €20 294 compared with grade II, and grade IV added €25 190 to €63 644 compared with grade III. Conclusion: Complications following pancreatoduodenectomy increase costs substantially, exceeding €100 000 for a single postoperative pancreatic fistula grade C. These costs can now be used to design cost-effective prophylactic and therapeutic interventions aiming to improve patient outcomes together with optimal use of healthcare resources.
- Open Access
- 22 June 2026
- Published by Oxford University Press (OUP) in British Journal of Surgery
- Vol. 113 (9)
Surgical Outcomes from Nationwide Implementation of the International Best-Practice for Locally Advanced Pancreatic Cancer (PREOPANC-4) study
Background: In expert centers, surgical resection rates of locally advanced pancreatic cancer (LAPC) following induction chemotherapy have increased beyond 20% with subsequent 25% five-year overall survival (OS). In the Netherlands, however, the historical low 8% LAPC resection rate compared with 23% in international expert centers reflects relative reluctance. Thereby, opportunities to achieve long-term survival in appropriately selected patients may be missed. This study evaluated whether nationwide implementation of international multidisciplinary best-practice for LAPC management is feasible while maintaining surgical safety benchmarks (in-hospital/30-day major morbidity <50% and mortality ≤5%). Methods: A multidisciplinary protocol was designed in collaboration with four international experts and prospectively implemented nationwide within the Dutch Pancreatic Cancer Group (DPCG) (2022-2024). This observational cohort included consecutive patients diagnosed with LAPC, defined by DPCG criteria. Eligible patients had radiologically non-progressive disease after at least four months of multi-agent chemotherapy. All patients who underwent resection were included in this safety analysis. A predefined sub-group analysis included patients with National Comprehensive Cancer Network (NCCN) LAPC. Primary outcomes included in-hospital/30-day major morbidity (i.e., Clavien-Dindo grade ≥IIIa) and mortality. The expected number of resections was 53. Results: Overall, 180 patients with LAPC underwent surgical exploration, of whom 155 (86%) underwent resection in 11 centers. Most (74%) resections were performed in the three LAPC surgical centers. Extended resections were performed in 77% of patients, including portomesenteric venous (60%), multivisceral (23%), and arterial (21%) resections. In-hospital/30-day major morbidity rate was 44% and mortality rate was 0.6%, both within pre-established safety benchmarks. Benchmarks were also reached for patients with NCCN LAPC (49% major morbidity, 2% mortality). Conclusion: Nationwide implementation of the international best-practice for LAPC was feasible with nearly three times more resections performed than expected, while morbidity and mortality remained well within predefined safety benchmarks.
- 22 June 2026
- Published by Oxford University Press (OUP) in British Journal of Surgery
- Vol. 113 (9)
Open AccessSurgical Outcomes from Nationwide Implementation of the International Best-Practice for Locally Advanced Pancreatic Cancer (PREOPANC-4) study
Background: In expert centers, surgical resection rates of locally advanced pancreatic cancer (LAPC) following induction chemotherapy have increased beyond 20% with subsequent 25% five-year overall survival (OS). In the Netherlands, however, the historical low 8% LAPC resection rate compared with 23% in international expert centers reflects relative reluctance. Thereby, opportunities to achieve long-term survival in appropriately selected patients may be missed. This study evaluated whether nationwide implementation of international multidisciplinary best-practice for LAPC management is feasible while maintaining surgical safety benchmarks (in-hospital/30-day major morbidity <50% and mortality ≤5%). Methods: A multidisciplinary protocol was designed in collaboration with four international experts and prospectively implemented nationwide within the Dutch Pancreatic Cancer Group (DPCG) (2022-2024). This observational cohort included consecutive patients diagnosed with LAPC, defined by DPCG criteria. Eligible patients had radiologically non-progressive disease after at least four months of multi-agent chemotherapy. All patients who underwent resection were included in this safety analysis. A predefined sub-group analysis included patients with National Comprehensive Cancer Network (NCCN) LAPC. Primary outcomes included in-hospital/30-day major morbidity (i.e., Clavien-Dindo grade ≥IIIa) and mortality. The expected number of resections was 53. Results: Overall, 180 patients with LAPC underwent surgical exploration, of whom 155 (86%) underwent resection in 11 centers. Most (74%) resections were performed in the three LAPC surgical centers. Extended resections were performed in 77% of patients, including portomesenteric venous (60%), multivisceral (23%), and arterial (21%) resections. In-hospital/30-day major morbidity rate was 44% and mortality rate was 0.6%, both within pre-established safety benchmarks. Benchmarks were also reached for patients with NCCN LAPC (49% major morbidity, 2% mortality). Conclusion: Nationwide implementation of the international best-practice for LAPC was feasible with nearly three times more resections performed than expected, while morbidity and mortality remained well within predefined safety benchmarks.
- Open Access
- 12 May 2026
- Published by Oxford University Press (OUP) in BJS Open
- Vol. 10 (3)
Continuity of care experienced by patients diagnosed with pancreatic cancer: nationwide multicentre study
- 12 May 2026
- Published by Oxford University Press (OUP) in BJS Open
- Vol. 10 (3)
Open AccessContinuity of care experienced by patients diagnosed with pancreatic cancer: nationwide multicentre study
- Open Access
- 21 February 2026
- Published by Elsevier in SSRN Electronic Journal
Survival Benefit of Recurrence-focused Treatment After Surgery for Pancreatic Ductal Adenocarcinoma: A Propensity Score Matched Analysis of Patients from a Dutch National and US Institutional Cohort
Background: It is unclear whether recurrence-focused treatment (RFT) improves survival in all patients with recurrent pancreatic ductal adenocarcinoma (PDAC). This study aimed (1) to evaluate post-recurrence survival (PRS) in patients treated with RFT versus best supportive care (BSC)-only, and (2) to assess potential effect modification by prognosis at recurrence.Methods: This observational cohort study included patients with local or distant recurrence after PDAC resection (2014-2022) at all Dutch pancreatic surgery centers and NYU Langone Health. Patients with unknown RFT or survival status were excluded. Prognosis at recurrence was defined using the PRS risk score, stratifying patients into worse (≥174.5) and better (<174.5) prognostic subgroups. Propensity score matching (PSM) was performed in the overall cohort, and within subgroups. Median PRS in RFT versus BSC-only patients was estimated and compared using Kaplan-Meier curve analysis and the log-rank test. Multivariable Cox regression was used to adjust for residual imbalance after PSM, with results presented as hazard ratios (HR) and 95% confidence intervals (95%CI).Findings: Overall, 1,423 patients were included. After PSM, 344 RFT patients were matched to 344 BSC-only patients. The multivariable analysis included ECOG performance score and country as a random effect. PRS was 10 (95%CI 9-12) months for RFT versus 4 (95%CI 3-4) months for BSC-only (adjusted HR0.38, 95%CI: 0.31-0.48, P<0.001). In the worse prognosis group, PRS was 10 (95%CI 9-11) versus 3 (95%CI 3-4) months (adjusted HR0.38, 95%CI: 0.30-0.49, P<0.001). In the better prognosis group, PRS was 18 (95%CI 12-27) months versus 8 (95%CI 4-14) months (adjusted HR0.24, 95%CI: 0.10-0.55, P<0.001).Interpretation: Treatment for PDAC recurrence was associated with significantly longer PRS compared with BSC-only, irrespective of prognosis at recurrence. The survival benefit associated with RFT supports its consideration in the management of patients with PDAC recurrence.
- 21 February 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessSurvival Benefit of Recurrence-focused Treatment After Surgery for Pancreatic Ductal Adenocarcinoma: A Propensity Score Matched Analysis of Patients from a Dutch National and US Institutional Cohort
Background: It is unclear whether recurrence-focused treatment (RFT) improves survival in all patients with recurrent pancreatic ductal adenocarcinoma (PDAC). This study aimed (1) to evaluate post-recurrence survival (PRS) in patients treated with RFT versus best supportive care (BSC)-only, and (2) to assess potential effect modification by prognosis at recurrence.Methods: This observational cohort study included patients with local or distant recurrence after PDAC resection (2014-2022) at all Dutch pancreatic surgery centers and NYU Langone Health. Patients with unknown RFT or survival status were excluded. Prognosis at recurrence was defined using the PRS risk score, stratifying patients into worse (≥174.5) and better (<174.5) prognostic subgroups. Propensity score matching (PSM) was performed in the overall cohort, and within subgroups. Median PRS in RFT versus BSC-only patients was estimated and compared using Kaplan-Meier curve analysis and the log-rank test. Multivariable Cox regression was used to adjust for residual imbalance after PSM, with results presented as hazard ratios (HR) and 95% confidence intervals (95%CI).Findings: Overall, 1,423 patients were included. After PSM, 344 RFT patients were matched to 344 BSC-only patients. The multivariable analysis included ECOG performance score and country as a random effect. PRS was 10 (95%CI 9-12) months for RFT versus 4 (95%CI 3-4) months for BSC-only (adjusted HR0.38, 95%CI: 0.31-0.48, P<0.001). In the worse prognosis group, PRS was 10 (95%CI 9-11) versus 3 (95%CI 3-4) months (adjusted HR0.38, 95%CI: 0.30-0.49, P<0.001). In the better prognosis group, PRS was 18 (95%CI 12-27) months versus 8 (95%CI 4-14) months (adjusted HR0.24, 95%CI: 0.10-0.55, P<0.001).Interpretation: Treatment for PDAC recurrence was associated with significantly longer PRS compared with BSC-only, irrespective of prognosis at recurrence. The survival benefit associated with RFT supports its consideration in the management of patients with PDAC recurrence.
- Open Access
The Value of Stereotactic Radiotherapy After FOLFIRINOX in Patients with Pancreatic Cancer with Vascular Contact—A Nationwide, Retrospective Cohort Study
Background/Objectives: Stereotactic body radiotherapy (SBRT) aims to prolong overall survival (OS) in patients with pancreatic ductal adenocarcinoma (PDAC) with vascular contact without progression of disease after (m)FOLFIRINOX. The primary objective of this study was to determine the potential value of SBRT. Methods: This nationwide, retrospective cohort study included patients with PDAC without progression of disease after at least four cycles of (m)FOLFIRINOX. The study comprised two cohorts, the SBRT and the No SBRT group. A landmark analysis excluded patients with a follow-up or OS time less than 12 months to minimize immortal time bias in the SBRT group. The primary outcome was OS from diagnosis. Secondary outcomes were the histopathological characteristics after resection. Results: Overall, 331 patients were included, of whom 231 were in the landmark analysis. In the overall cohort, the median OS was 20.7 months in the SBRT group versus 15.7 months in the No SBRT group (p = 0.004). In the landmark analysis, the median OS was 23.2 months in the SBRT group compared to 22.3 months in the No SBRT group (p = 0.554). These results indicate the presence of immortal time bias in the overall cohort in favor of the SBRT group. In the subgroup after resection, ypT0-2 (95% versus 76.5% [p = 0.026]), ypN0 (75% versus 37.3% [p < 0.004]), and absence of perineural invasion (50% versus 68.6% [p = 0.015]) were more prevalent in the SBRT group. Conclusions: In a landmark analysis, including only patients who survived at least 12 months after diagnosis, we found no difference in median OS between (m)FOLFIRINOX-only and (m)FOLFIRINOX with consecutive SBRT.
Open AccessThe Value of Stereotactic Radiotherapy After FOLFIRINOX in Patients with Pancreatic Cancer with Vascular Contact—A Nationwide, Retrospective Cohort Study
Background/Objectives: Stereotactic body radiotherapy (SBRT) aims to prolong overall survival (OS) in patients with pancreatic ductal adenocarcinoma (PDAC) with vascular contact without progression of disease after (m)FOLFIRINOX. The primary objective of this study was to determine the potential value of SBRT. Methods: This nationwide, retrospective cohort study included patients with PDAC without progression of disease after at least four cycles of (m)FOLFIRINOX. The study comprised two cohorts, the SBRT and the No SBRT group. A landmark analysis excluded patients with a follow-up or OS time less than 12 months to minimize immortal time bias in the SBRT group. The primary outcome was OS from diagnosis. Secondary outcomes were the histopathological characteristics after resection. Results: Overall, 331 patients were included, of whom 231 were in the landmark analysis. In the overall cohort, the median OS was 20.7 months in the SBRT group versus 15.7 months in the No SBRT group (p = 0.004). In the landmark analysis, the median OS was 23.2 months in the SBRT group compared to 22.3 months in the No SBRT group (p = 0.554). These results indicate the presence of immortal time bias in the overall cohort in favor of the SBRT group. In the subgroup after resection, ypT0-2 (95% versus 76.5% [p = 0.026]), ypN0 (75% versus 37.3% [p < 0.004]), and absence of perineural invasion (50% versus 68.6% [p = 0.015]) were more prevalent in the SBRT group. Conclusions: In a landmark analysis, including only patients who survived at least 12 months after diagnosis, we found no difference in median OS between (m)FOLFIRINOX-only and (m)FOLFIRINOX with consecutive SBRT.