Key Points
Question
Is pancreatic enucleation a safe alternative to pancreatoduodenectomy for benign and low-grade malignant tumors of the pancreatic head?
Findings
In this multicenter cohort study of 660 patients (217 propensity-matched pairs), enucleation was associated with higher clinically relevant pancreatic fistula (33.2% vs 10.6%) but fewer major complications (8.8% vs 17.1%), less delayed gastric emptying, shorter hospitalization, and superior long-term quality of life. Enucleation’s advantage was greatest for tumors 2.5 cm or smaller.
Meaning
Enucleation should be considered a viable option for benign and low-grade pancreatic head tumors when technically feasible, particularly for tumors 2.5 cm or smaller.
This cohort study of patients with benign or low-grade pancreatic head or uncinate process tumors assesses short-term surgical outcomes and long-term quality of life after enucleation vs pancreatoduodenectomy.
Abstract
Importance
Pancreatic enucleation may offer a parenchyma-sparing alternative to pancreatoduodenectomy for benign and low-grade malignant pancreatic head tumors, but its perioperative risk-benefit profile relative to pancreatoduodenectomy remains incompletely defined.
Objective
To compare short-term outcomes and long-term quality of life after enucleation vs pancreatoduodenectomy for benign and low-grade malignant tumors of the pancreatic head and uncinate process.
Design, Setting, and Participants
This multicenter, retrospective cohort study assessed patients undergoing enucleation or pancreatoduodenectomy for benign or low-grade pancreatic head or uncinate process tumors at 4 high-volume Chinese centers (January 1, 2015, to December 31, 2024; median follow-up, 36 months) with 1:1 propensity score matching. Data analysis was performed from March to May 2026.
Exposure
Pancreatic enucleation vs pancreatoduodenectomy.
Main Outcomes and Measures
The primary end point was clinically relevant postoperative pancreatic fistula (CR-POPF; International Study Group on Pancreatic Surgery grade B or C). Secondary end points included major complications (Clavien-Dindo grade III or higher), delayed gastric emptying, length of stay, long-term quality of life (European Organisation for Research and Treatment of Cancer Quality of Life Core Questionnaire and 26-item Quality of Life Pancreatic Cancer Questionnaire), and new-onset diabetes. Tumor-to-main pancreatic duct distance was evaluated through stratified and multivariable analyses.
Results
Of 660 patients (297 undergoing enucleation and 363 undergoing pancreatoduodenectomy; 355 [53.8%] female; median [IQR] age, 55.0 [41.0-65.0] years), 217 matched pairs were formed. CR-POPF was higher after enucleation (72 [33.2%] vs 23 [10.6%]; adjusted odds ratio, 4.22; 95% CI, 2.47-7.21; P < .001). Major complications were lower (19 [8.8%] vs 37 [17.1%]; adjusted odds ratio, 0.37; 95% CI, 0.20-0.70; P = .002) as was delayed gastric emptying (10 [4.6%] vs 55 [25.3%]; P < .001). Median (IQR) stay was 9 (6-14) vs 15 (11-22) days (P < .001). A total of 551 patients (83.5%) completed quality-of-life assessment; enucleation was associated with higher global health status (mean [SD] scores, 77.4 [13.7] vs 62.1 [18.0]; P < .001), lower new-onset diabetes (7.8% [20 of 257] vs 33.3% [98 of 294]; P < .001), and higher return to work (92.7% [191 of 206] vs 72.0% [121 of 168]; P < .001). Enucleation’s complication advantage was greatest for tumors 2.5 cm or smaller (5.8% [6 of 104 patients] vs 26.2% [17 of 65 patients]; P for interaction = .01).
Conclusions and Relevance
In this multicenter cohort study, enucleation for benign and low-grade pancreatic head tumors was associated with higher CR-POPF than pancreatoduodenectomy but fewer major complications, less delayed gastric emptying, shorter hospitalization, and superior long-term quality of life. These findings support enucleation as a viable option when technically feasible, particularly for tumors 2.5 cm or smaller; residual tumor size imbalance after matching warrants prospective confirmation.
Introduction
Pancreatic benign and low-grade malignant tumors, including intraductal papillary mucinous neoplasms (IPMNs), serous cystadenomas, solid pseudopapillary neoplasms, and nonfunctional pancreatic neuroendocrine tumors, are increasingly detected, with pancreatic cystic neoplasms identified in approximately 2% to 3% of abdominal imaging studies. Unlike pancreatic ductal adenocarcinoma, these tumors carry a favorable prognosis after complete resection, with 5-year survival exceeding 90% for most subtypes. However, the indication for surgical resection remains debated because treatment must balance oncologic adequacy against the morbidity of pancreatic surgery.
For tumors in the pancreatic head, pancreatoduodenectomy has long been the standard surgical approach, but it carries considerable perioperative morbidity, including postoperative pancreatic fistula (POPF), postpancreatectomy hemorrhage, and delayed gastric emptying (DGE), with complication rates of 30% to 50%. Long-term functional sequelae of pancreatoduodenectomy, particularly new-onset diabetes and exocrine insufficiency, are of concern for patients with benign or low-grade tumors who have a long life expectancy. Pancreatic enucleation (local excavation of a tumor without disrupting pancreatic ductal continuity or gastrointestinal reconstruction) has emerged as a parenchyma-sparing alternative that preserves endocrine and exocrine function. Recent evidence supports favorable long-term outcomes after enucleation. Giuliani et al reported quality-of-life scores comparable to the general population at a median follow-up of 74 months, with new-onset diabetes in only 7.1%, and Li et al demonstrated preserved endocrine and exocrine function even when main pancreatic duct (MPD) repair was required during minimally invasive enucleation. However, concerns persist regarding higher POPF rates after enucleation, particularly for tumors close to the MPD, and the inability to perform regional lymphadenectomy remains a consideration for lesions with uncertain malignant potential.
Despite increasing interest in organ-preserving surgery, comparative studies of enucleation vs pancreatoduodenectomy for benign and low-grade tumors of the pancreatic head and uncinate process remain limited, constrained by small sample sizes, single-arm designs, or heterogeneous inclusion criteria. Few have used propensity score matching to address the inherent selection bias whereby patients with smaller, more favorably located tumors are preferentially selected for enucleation. Notably, to our knowledge, no study to date has evaluated whether the comparative safety of enucleation vs pancreatoduodenectomy varies according to tumor to MPD distance—an anatomical parameter that fundamentally influences both surgical feasibility and complication risk. We therefore compared short-term surgical outcomes and long-term quality of life after enucleation vs pancreatoduodenectomy for benign and low-grade malignant tumors of the pancreatic head and uncinate process.
Methods
Design and Participants
This multicenter, retrospective cohort study included patients who underwent enucleation or pancreatoduodenectomy for benign or low-grade malignant tumors of the pancreatic head and uncinate process at 4 high-volume institutions in China (Chinese People’s Liberation Army [PLA] General Hospital, Beijing; the Affiliated Hospital of Qingdao University, Qingdao; the First Hospital of Jilin University, Changchun; and the First Affiliated Hospital of Jinzhou Medical University, Jinzhou) between January 1, 2015, and December 31, 2024. The 4 centers form a pancreatic surgery collaborative network coordinated by the Chinese PLA General Hospital, with surgeons across the institutions sharing a common pancreatic surgery training background. Each institution is a tertiary, high-volume center that routinely performs both enucleation and pancreatoduodenectomy, so procedure allocation reflected intraoperative surgical indication rather than center-level procedural availability. Consecutive eligible patients were enrolled at each site. The study was approved by the Institutional Ethics Committee of Chinese PLA General Hospital; the requirement for informed consent was waived owing to the retrospective design. This study was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for cohort studies.
Inclusion criteria were as follows: (1) single histologically confirmed benign or low-grade malignant tumor (nonfunctional pancreatic neuroendocrine tumor [World Health Organization grade 1 to 2], solid pseudopapillary neoplasm, branch-duct IPMN, serous cystic neoplasm, or mucinous cystic neoplasm); (2) tumor location in the pancreatic head or uncinate process; (3) no evidence of MPD or common bile duct invasion on preoperative imaging; and (4) available preoperative cross-sectional imaging. Key exclusion criteria included pancreatic ductal adenocarcinoma or other high-grade malignant tumor, multifocal tumors, prior pancreatic resection, insufficient preoperative imaging for tumor to MPD distance assessment, and substantial cardiopulmonary comorbidity precluding major abdominal surgery (American Society of Anesthesiologists [ASA] class IV or V); the complete list and surgical procedures are described in the eMethods in Supplement 1. The choice between enucleation and pancreatoduodenectomy was determined intraoperatively by the attending surgeon according to standardized criteria applied uniformly across all centers and informed by intraoperative ultrasonography based on tumor size and location, proximity to the MPD, and the patient’s overall tolerance. Both procedures were performed via robotic, laparoscopic, and open approaches, with minimally invasive approaches predominating. Standardized surgical and perioperative protocols are described in the eMethods in Supplement 1. Sex data were recorded as male or female based on self-report at hospital admission. Race or ethnicity data were not systematically collected; the study population was exclusively Chinese.
Outcomes
The primary end point was clinically relevant POPF (CR-POPF), defined as International Study Group on Pancreatic Surgery grade B or C according to the 2016 criteria. Secondary end points included major complications (Clavien-Dindo grade III or higher), postpancreatectomy hemorrhage, DGE, bile leak, postoperative length of stay, 30-day and 90-day mortality, and long-term outcomes, including quality of life, new-onset diabetes, and pancreatic exocrine insufficiency. Tumor to MPD distance was measured on preoperative contrast-enhanced computed tomography or magnetic resonance imaging, and patients were classified into 3 strata (≤2, 2-3, and >3 mm) according to previously reported thresholds. To address selection bias, we performed 1:1 propensity score matching using multivariable logistic regression, including age, sex, body mass index, ASA classification, tumor size, diabetes, hypertension, and tumor pathologic type, with nearest-neighbor matching without replacement at a caliper of 0.2 SDs of the logit. Covariate balance was assessed using the absolute standardized mean difference (SMD), with SMD less than 0.10 considered adequate.
Statistical Analysis
In the matched cohort, continuous variables were compared using the Wilcoxon signed-rank test and binary outcomes using McNemar test or conditional logistic regression. Multivariable logistic regression was used to identify factors independently associated with CR-POPF and major complications, with results reported as adjusted odds ratios (AORs) with 95% CIs. Multiplicative interaction terms were used to assess effect modification by tumor to MPD distance and tumor size. All tests were 2-tailed with P < .05 considered statistically significant; no multiplicity adjustment was applied to secondary end point analyses. Quality of life was assessed at a single cross-sectional time point using the European Organisation for Research and Treatment of Cancer (EORTC) Quality of Life Core Questionnaire (QLQ-C30) (version 3.0) and the 26-item Quality of Life Pancreatic Cancer Questionnaire (QLQ-PAN26) module, with scores linearly transformed to a 0- to 100-point scale per the EORTC scoring manual. For functional scales, higher scores indicate better function; for symptom scales, higher scores indicate greater symptom burden. A worst-case sensitivity analysis was performed to assess the impact of loss to follow-up. Additional methodologic details are provided in the eMethods in Supplement 1. Data analysis was performed from March to May 2026.
Results
A total of 660 consecutive patients (355 [53.8%] female and 305 [46.2%] male; median [IQR] age, 55.0 [41.0-65.0] years) were identified (297 undergoing enucleation and 363 undergoing pancreatoduodenectomy) (eFigure 1 in Supplement 1). Before propensity score matching (eTable 1 and eFigure 2 in Supplement 1), patients in the pancreatoduodenectomy group were older (median [IQR] age, 62.0 [53.0-69.5] vs 51.0 [37.0-60.0] years), more likely to be male (196 [54.0%] vs 109 [36.7%]), and had larger tumors (median [IQR], 3.5 [2.5-5.0] vs 2.5 [1.8-4.0] cm) (P < .001 for all). Pathologic type distribution differed substantially between groups (Table 1). After 1:1 matching, 217 pairs (434 patients) were formed with adequate covariate balance (all SMDs <0.20 except tumor size [SMD = 0.31]) (eFigure 3 and eTable 2 in Supplement 1). A doubly robust sensitivity analysis confirmed robustness to this residual imbalance (eTable 3 in Supplement 1). Baseline characteristics are summarized in Table 1.
Table 1. Baseline Characteristics of the Study Cohort Before Propensity Score Matching.
| Characteristic | No. (%) of patientsa | |
|---|---|---|
| Enucleation (n = 297) | Pancreatoduodenectomy (n = 363) | |
| Age, median (IQR), y | 51 (37-60) | 62 (53-70) |
| Demographics | ||
| Sex | ||
| Male | 109 (26.7) | 196 (54.0) |
| Female | 188 (63.3) | 167 (46.0) |
| BMI, mean (SD) | 24.4 (3.9) | 23.3 (3.6) |
| ASA class ≥3 | 17 (5.7) | 56 (15.4) |
| Preoperative diabetes | 32 (10.8) | 63 (16.9) |
| Tumor characteristics | ||
| Tumor location | ||
| Pancreatic head | 136 (45.8) | 357 (98.3) |
| Pancreatic neck | 67 (22.6) | 0 |
| Head or uncinate | 44 (14.8) | 0 |
| Uncinate | 22 (7.4) | 4 (1.1) |
| Head-neck or body | 23 (7.7) | 0 |
| Other | 5 (1.7) | 2 (0.6) |
| Tumor size, median (IQR), cm | 2.5 (1.8-4.0) | 3.5 (2.5-5.0) |
| Histopathologic findings | ||
| IPMN | 61 (20.5) | 224 (61.7) |
| SPN | 86 (29.0) | 43 (11.8) |
| SCN | 38 (12.8) | 17 (4.7) |
| pNET | 74 (24.9) | 39 (10.7) |
| MCN | 9 (3.0) | 8 (2.2) |
| Otherb | 29 (9.8) | 32 (8.8) |
| MPD distance, median (IQR), mm | 4.2 (2.4-5.8) | 1.6 (0.0-2.8) |
Abbreviations: ASA, American Society of Anesthesiologists; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); IPMN, intraductal papillary mucinous neoplasm; MCN, mucinous cystic neoplasm; MPD, main pancreatic duct; pNET, pancreatic neuroendocrine tumor; SCN, serous cystic neoplasm; SPN, solid pseudopapillary neoplasm.
Unless otherwise indicated.
Other indicates patients with multifocal tumors.
In the matched cohort, enucleation was associated with shorter operative time and less blood loss (eTables 4 and 5 in Supplement 1). The primary end point, CR-POPF (grade B or C), occurred in 72 of the 217 patients undergoing enucleation (33.2%) vs 23 of the 217 patients undergoing pancreatoduodenectomy (10.6%) (P < .001), driven predominantly by grade B fistula (grade C: 4 [1.8%] vs 1 [0.5%]) (eFigure 4 and eTable 6 in Supplement 1). Management reflected fistula severity: grade B fistulas were managed conservatively with prolonged drainage, supplemented by percutaneous drainage for fluid collections developing after drain removal, whereas grade C fistulas required interventional management (eResults in Supplement 1). No POPF-related deaths occurred in the enucleation group. However, major complications (Clavien-Dindo grade III or higher) were significantly lower after enucleation (19 [8.8%] vs 37 [17.1%]; P = .02) (eFigure 5 and eTable 7 in Supplement 1) as was DGE (10 [4.6%] vs 55 [25.3%]; P < .001). Bile leak was more frequent after pancreatoduodenectomy (29 [13.4%] vs 3 [1.4%]; P < .001). Postpancreatectomy hemorrhage, intra-abdominal abscess, and reoperation rates were comparable between groups (eTable 5 in Supplement 1). Median (IQR) postoperative length of stay was 9 (6-14) days after enucleation vs 15 (11-22) days after pancreatoduodenectomy (P < .001). Perioperative mortality was zero in the enucleation group (eTable 5 in Supplement 1); in the pancreatoduodenectomy group, 1 patient died on postoperative day 5 from cardiac arrest.
On multivariable logistic regression, enucleation was an independent risk factor for CR-POPF (AOR, 4.22; 95% CI, 2.47-7.21; P < .001) but an independent protective factor against major complications (AOR, 0.37; 95% CI, 0.20-0.70; P = .002) (eTable 8 and eFigure 6 in Supplement 1). In stratified analysis (Table 2; eFigure 7 and eTable 9 in Supplement 1), CR-POPF was consistently higher after enucleation across all distance strata (P for interaction = .74). For major complications, enucleation was noninferior at a tumor to MPD distance of 2 mm or smaller (4 of 46 [8.7%] vs 37 of 215 [17.2%]; P = .18), had a nonstatistically significant association with lower rates at a tumor to MPD distance of 2 to 3 mm (8 of 74 [10.8%] vs 14 of 60 [23.3%]; P = .09), and was significantly protective at a tumor to MPD distance larger than 3 mm (10 of 168 [6.0%] vs 17 of 71 [16.9%]; P = .02). On multivariable regression, enucleation remained independently associated with CR-POPF, whereas MPD distance itself was not significant (eTable 8 in Supplement 1). The enucleation group had a significantly greater median (IQR) tumor to MPD distance than the pancreatoduodenectomy group (4.2 [2.4-5.8] vs 1.6 [0.0-2.8] mm; P < .001), reflecting inherent selection bias in surgical decision-making; this difference persisted after matching (4.3 vs 1.0 mm). In the 2 mm or smaller tumor to MPD distance stratum, CR-POPF was found in 17 of 46 patients undergoing enucleation (37.0%) vs 30 of 215 undergoing pancreatoduodenectomy (14.0%) (P = .001), and median (IQR) length of stay was 10 (7-16) vs 17 (12-24) days in the enucleation vs pancreatoduodenectomy group (P < .001). In the 2- to 3-mm tumor to MPD distance stratum, CR-POPF was found in 29 of 74 patients undergoing enucleation (39.2%) vs 12 of 60 patients undergoing pancreatoduodenectomy (20.0%) (P = .03), and median (IQR) length of stay was 10 (7-16) vs 17 (11-22) days (P < .001). In the greater than 3-mm tumor to MPD distance stratum, CR-POPF was found in 53 of 168 patients undergoing enucleation (31.5%) vs 12 of 71 patients undergoing pancreatoduodenectomy (16.9%) (P = .03), and median (IQR) length of stay was 9 (6-15) vs 17 (14-23) days (P < .001). Exploratory analysis identified 2.5 cm as the threshold at which the major complication advantage of enucleation was most pronounced (P for interaction = .01) (Figure 1): for tumors 2.5 cm or smaller, major complications were markedly lower after enucleation (5.8% [6 of 104 patients] vs 26.2% [17 of 65 patients]; P < .001), with the benefit extending to tumors 2 cm or smaller (1 of 88 [1.1%] vs 8 of 61 [13.1%]; P = .004) (eFigure 9 and eTable 13 in Supplement 1).
Table 2. Outcomes Stratified by Tumor to MPD Distance in 634 Patientsa.
| Outcome | Enucleation | Pancreatoduodenectomy | P valueb |
|---|---|---|---|
| MPD distance in the full cohort, median (IQR), mm | 4.2 (2.4-5.8) | 1.6 (0.0-2.8) | <.001 |
| Tumor to MPD distance stratum | |||
| ≤2 mm | |||
| CD grade ≥III, No./total No. (%) | 4/46 (8.7) | 37/215 (17.2) | .18 |
| POPF grade B or C, No./total No. (%) | 17/46 (37.0) | 30/215 (14.0) | .001 |
| LOS, median (IQR), d | 10 (7-16) | 17 (12-24) | <.001 |
| 2-3 mm | |||
| CD grade ≥III, No./total No. (%) | 8/74 (10.8) | 14/60 (23.3) | .09 |
| POPF grade B or C, No./total No. (%) | 29/74 (39.2) | 12/60 (20.0) | .023 |
| LOS, median (IQR), d | 10 (7-16) | 17 (11-22) | <.001 |
| >3 mm | |||
| CD grade ≥III, No./total No. (%) | 10/168 (6.0%) | 12/71 (16.9) | .02 |
| POPF grade B or C, No./total No. (%) | 53/168 (31.5) | 12/71 (16.9) | .03 |
| LOS, median (IQR), d | 9 (6-15) | 17 (14-23) | <.001 |
| Risk summary | |||
| CD grade ≥III, No./total No. (%) | |||
| ≤3 mm (High risk) | 12/120 (10.0) | 51/275 (18.5) | .047 |
| >3 mm (Low risk) | 10/168 (6.0) | 12/71 (16.9) | .02 |
| POPF grade B or C, No./total No. (%) | |||
| ≤3 mm (High risk) | 46/120 (38.3) | 42/275 (15.3) | <.001 |
| >3 mm (Low risk) | 53/168 (31.5) | 12/71 (16.9) | .03 |
| LOS, median (IQR), d | |||
| ≤3 mm (High risk) | 10 (7-16) | 17 (12-23) | <.001 |
| >3 mm (Low risk) | 9 (6-15) | 17 (14-23) | <.001 |
Abbreviations: CD, Clavien-Dindo; LOS, length of stay; MPD, main pancreatic duct; POPF, postoperative pancreatic fistula.
Tumor to MPD distance was measured on preoperative contrast-enhanced computed tomography or magnetic resonance imaging and classified as 2 mm or less (high risk), 2 to 3 mm (intermediate risk), and greater than 3 mm (low risk).
P values were derived from the χ2 test or Fisher exact test for categorical outcomes and the Mann-Whitney U test for length of stay.
Figure 1. Line Graphs and Dot Plot Showing Outcomes Stratified by Tumor Size at the 2.5-cm Threshold .
Major complication and clinically relevant postoperative pancreatic fistula (CR-POPF) rates are shown for enucleation vs pancreatoduodenectomy stratified by tumor size. The 2.5-cm threshold was identified by exploratory interaction analysis (P for interaction = .01) as the point at which the between-group difference in major complications was most pronounced. For tumors 2.5 cm or smaller, enucleation was associated with markedly lower major complications (5.8% vs 26.2%); for tumors larger than 2.5 cm, rates were comparable (11.5% vs 13.2%). Errors bars indicate 95% CIs; diamond, overall pooled effect size. CD indicates Clavien-Dindo; OR, odds ratio.
In the full cohort, quality of life was assessed in 551 of 660 patients (83.5%) at 36 months of follow-up; a total of 109 patients (16.5%) were lost to follow-up (Figure 2). Enucleation was associated with significantly better global health status (mean [SD] score, 77.4 [13.7] vs 62.1 [18.0]; P < .001) (Table 3; eTable 10 and eFigure 8 in Supplement 1). New-onset diabetes was lower after enucleation (20 of 257 [7.8%] vs 98 of 294 [33.3%]; P < .001), and return-to-work rate was higher (191 of 206 [92.7%] vs 121 of 168 [72.0%]; P < .001). The only domain less favorable after enucleation was future worry on QLQ-PAN26 (mean [SD], 27.3 [20.6] vs 20.5 [16.6]; P < .001), likely reflecting patient anxiety about whether parenchyma-sparing resection had adequately removed all tumor tissue compared with a formal resection with lymphadenectomy. All significant differences persisted in the propensity score–matched cohort (eTable 11 in Supplement 1) and a worst-case sensitivity analysis (eTable 12 in Supplement 1). Subgroup analyses by tumor size and pathologic type are presented in eFigures 9 to 11 and eTables 13 and 14 in Supplement 1.
Figure 2. Circle and Line Graphs Showing Long-Term Quality of Life After Enucleation vs Pancreatoduodenectomy.
Mean European Organisation for Research and Treatment of Cancer Quality of Life Core Questionnaire functional and symptom scale scores and selected and 26-item Quality of Life Pancreatic Cancer Questionnaire (QLQ-PAN26) items are shown for enucleation vs pancreatoduodenectomy at a median follow-up of 36 months. Higher functional scale scores and lower symptom scale scores indicate better quality of life (QoL). Enucleation was associated with significantly higher global health status (77.4 vs 62.1), physical function (92.5 vs 71.3), role function, emotional function, and social function and lower fatigue (15.6 vs 32.1) and pain (8.8 vs 23.6) (P < .001 for all). QoL was assessed in 551 of 660 patients (83.5%).
Table 3. Long-Term Quality of Life, Endocrine and Exocrine Function, and Social Recoverya.
| Domain | Enucleation (n = 257) | Pancreatoduodenectomy (n = 294) | P valueb |
|---|---|---|---|
| Follow-up, median (IQR), mo | 36 (24-54) | 35 (24-49) | .27 |
| EORTC QLQ-C30 functional scale scores | |||
| Global health status | 77.4 (13.7) | 62.1 (18.0) | <.001 |
| Physical function | 92.5 (7.9) | 71.3 (15.5) | <.001 |
| Role function | 87.3 (11.4) | 66.7 (20.0) | <.001 |
| Social function | 86.9 (11.6) | 66.0 (19.6) | <.001 |
| EORTC QLQ-C30 symptom scale scores | |||
| Fatigue | 15.6 (13.7) | 32.1 (19.9) | <.001 |
| Nausea and vomiting | 5.9 (6.8) | 14.2 (13.0) | <.001 |
| Pain | 8.8 (8.7) | 23.6 (16.9) | <.001 |
| Appetite loss | 5.2 (6.6) | 18.4 (15.1) | <.001 |
| Diarrhea | 10.3 (12.1) | 28.2 (19.1) | <.001 |
| Constipation | 9.1 (10.2) | 16.0 (13.9) | <.001 |
| EORTC QLQ-PAN26 scores | |||
| Abdominal pain | 9.4 (9.3) | 25.1 (18.9) | <.001 |
| Bloating | 16.6 (15.8) | 29.3 (19.1) | <.001 |
| Indigestion | 11.6 (11.6) | 27.9 (21.2) | <.001 |
| Bowel changes | 9.9 (10.1) | 30.4 (22.0) | <.001 |
| Steatorrhea | 6.1 (7.0) | 21.5 (17.2) | <.001 |
| Body image | 21.1 (16.2) | 26.3 (18.9) | .002 |
| Sexual function | 16.5 (15.5) | 37.1 (24.7) | <.001 |
| Postprandial discomfort | 10.6 (10.5) | 24.6 (18.8) | <.001 |
| Future worry | 27.3 (20.6) | 20.5 (16.6) | <.001 |
| Endocrine and exocrine function, No./total No. (%) | |||
| New-onset diabetes | 20/257 (7.8) | 98/294 (33.3) | <.001 |
| Pancreatic enzyme replacement | 16/257 (6.2) | 100/294 (34.0) | <.001 |
| Clinical steatorrhea | 18/257 (7.0) | 74/294 (25.2) | <.001 |
| Nutritional status | |||
| Current BMI | 23.0 (3.2) | 21.8 (2.8) | <.001 |
| Current albumin, g/dL | 4.32 (0.37) | 3.86 (0.37) | <.001 |
| ECOG performance status | 0.5 (0.7) | 1.3 (1.1) | <.001 |
| Social and functional recovery | |||
| Return to work, No./total No. (%) | 191/206 (92.7) | 121/168 (72.0) | <.001 |
| Time to return to work, mo | 2.4 (1.3) | 4.4 (2.3) | <.001 |
| Treatment satisfaction (1-5) | 4.2 (0.9) | 3.8 (1.1) | <.001 |
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); ECOG, Eastern Cooperative Oncology Group; EORTC QLQ-C30, European Organisation for Research and Treatment of Cancer Quality of Life Core Questionnaire; EORTC QLQ-PAN26, European Organisation for Research and Treatment of Cancer 26-item Quality of Life Pancreatic Cancer Questionnaire.
SI conversion factor: To convert albumin to grams per liter, multiply by 10.
Data are mean (SD) unless otherwise indicated. Quality of life was assessed using the EORTC QLQ-C30 (version 3.0) and QLQ-PAN26 module; all scores were linearly transformed to a 0- to 100-point scale per the EORTC scoring manual. For functional scales, higher scores indicate better function; for symptom scales, higher scores indicate greater symptom burden. Treatment satisfaction was measured on a 1- to 5-point Likert scale, with 1 indicating very dissatisfied and 5 indicating very satisfied. Assessments were completed in 551 of 660 patients (83.5%).
P values were derived from the Mann-Whitney U test for continuous variables and Fisher exact test for binary outcomes.
Discussion
The finding that enucleation is an independent risk factor for CR-POPF but a protective factor against major complications represents the central paradox of this procedure. Heidsma et al reported higher POPF grade B or C after enucleation vs formal pancreatectomy for pancreatic neuroendocrine tumors (24.5% vs 14.0%, P = .05), and Ou et al reported an overall CR-POPF rate of 50.0% in 146 enucleations, yet no grade C POPFs or 90-day mortality in the robotic subgroup. This paradox reflects fundamentally different pathophysiologic mechanisms: enucleation creates a parenchymal surface leak that is typically self-limited (predominantly grade B), whereas CR-POPF after pancreatoduodenectomy can be more aggressive due to activation of pancreatic enzymes by bile and intestinal fluids, often culminating in hemorrhage or abscess requiring intervention. In our series, the clinical trajectory of enucleation-associated POPF was markedly less severe—no perioperative deaths occurred in the enucleation group
The markedly lower incidence of DGE after enucleation (4.6% vs 25.3%) is a clinically important finding. DGE is one of the most debilitating complications after pancreatoduodenectomy, with reported incidences of 15% to 45%. The mechanism is multifactorial, including disruption of duodenal pacemaker activity and perianastomotic inflammation from pancreatic leak. Enucleation, by preserving the duodenum and avoiding gastrointestinal reconstruction, circumvents these mechanisms entirely. Bile leak was also more frequent after PD (13.4% vs 1.4%), attributable to hepaticojejunostomy in patients with typically nondilated bile ducts—a technical challenge that enucleation obviates. The long-term functional advantage of enucleation is well documented. Giuliani et al reported quality-of-life scores comparable to the general population at 74 months, with new-onset diabetes in only 7.1% and exocrine insufficiency in 10.0%. These data contrast sharply with the metabolic consequences of PD: exocrine insufficiency in 55% and endocrine insufficiency in 19%. Our findings (551 of 660 patients [83.5%] at follow-up) are consistent, demonstrating markedly lower new-onset diabetes (7.8% vs 33.3%) and higher return-to-work rates (92.7% vs 72.0%) after enucleation. The sole quality-of-life domain less favorable after enucleation (future worry) underscores the importance of comprehensive preoperative counseling.
Oncologic outcomes further supported the safety of parenchyma-sparing surgery. Among patients with low- to moderate-invasion tumors, recurrence, perioperative mortality, and nodal involvement were comparable after enucleation and pancreatoduodenectomy; for high-invasion tumors, the small number of enucleation cases precludes firm conclusions, and oncologic outcomes await confirmation in randomized clinical trials (eTable 16 in Supplement 1).
Our stratified analysis by tumor to MPD distance challenges the traditional view that proximity to the MPD (≤2-3 mm) is a relative contraindication to enucleation. Li et al reported preserved long-term endocrine and exocrine function even when MPD exposure, repair, or reconstruction was required during minimally invasive enucleation. Tang et al demonstrated safe robotic enucleation for tumors smaller than 3 mm from the MPD, and Ou et al showed no outcome differences by MPD distance at experienced centers. MPD repair (primary suture with or without stent) has been standardized by current consensus guidelines. On multivariable regression, enucleation remained independently associated with CR-POPF, whereas MPD distance itself was not significant, suggesting that surgical approach and MPD management ability exert a stronger influence on outcomes than anatomical distance per se. Critically, across all 3 tumor to MPD distance strata (≤2, 2-3, and >3 mm), enucleation was never associated with higher major complications (Table 2; eFigure 7 in Supplement 1). These findings suggest that tumor to MPD distance should not be used as the sole criterion to preclude enucleation at experienced centers. Exploratory analysis identified 2.5 cm as the tumor size threshold at which the enucleation advantage in reducing major complications is most pronounced (P for interaction = .01). Below 2.5 cm, enucleation reduces major complications by approximately 20 percentage points, a benefit that substantially outweighs the approximately 14–percentage point increase in CR-POPF (predominantly grade B). The international consensus guidelines for IPMN similarly suggest limited resection for small branch-duct lesions without high-risk stigmata. Above 2.5 cm, the major complication advantage disappears, whereas the CR-POPF excess widens; however, even above this threshold enucleation was not associated with higher major complications, and 2.5 cm should therefore guide shared decision-making rather than serve as an absolute contraindication. Among patients with tumors 2 cm or smaller, enucleation was associated with strikingly low rates of major complications (1.1%) and no DGE, with comparable CR-POPF to pancreatoduodenectomy. This finding is biologically plausible: smaller tumors are more likely to be superficially located and farther from the MPD, making them anatomically more amenable to complete excision without ductal injury. Our data provide quantitative evidence supporting a tumor size of 2 cm or smaller as a reasonable indication for enucleation when technically feasible.
In experienced high-volume centers with capacity for pancreatic fistula management, enucleation represents a reasonable parenchyma-sparing option in selected patients. With robotic assistance, intraoperative ultrasonography guidance, and standardized MPD repair techniques, enucleation can be safely performed even for tumors close to the MPD, although adoption in lower-volume centers should be phased (eTable 15 and eFigure 12 in Supplement 1). The high CR-POPF rate associated with enucleation requires institutional capacity for fistula management, including interventional radiology and endoscopic drainage. In lower-volume centers without such resources, the risk-benefit profile may differ, and a learning curve effect should be anticipated.
Limitations
This study has several limitations. First, despite propensity score matching, residual confounding may persist; tumor to MPD distance was not included in the matching model because it fundamentally influences surgical decision-making and the distance distributions remained different between groups after matching. However, this imbalance reflects appropriate clinical decision-making rather than inadequate matching because surgeons naturally favor enucleation for tumors farther from the MPD. The stratified and multivariable regression analyses provide complementary evidence, but these findings should be considered hypothesis-generating. Second, tumor to MPD distance was measured retrospectively on preoperative imaging; prospective measurement using standardized protocols would strengthen future studies. Third, quality of life was assessed at a single cross-sectional time point with 16.5% of patients lost to follow-up. However, a worst-case sensitivity analysis confirmed that the direction and significance of between-group differences were preserved across all key domains (eTable 12 in Supplement 1), and a doubly robust analysis confirmed that the findings were not attributable to residual tumor size imbalance (eTable 3 in Supplement 1). Fourth, the median postoperative length of stay in our cohort (15 days) was longer than reported in recent European series (7-11 days), reflecting differences in health care system structure and practice patterns between China and Western countries rather than inferior perioperative outcomes (eDiscussion in Supplement 1).
Conclusions
This study found that enucleation for benign and low-grade malignant tumors of the pancreatic head and uncinate process was associated with higher CR-POPF than pancreatoduodenectomy (predominantly grade B) but significantly fewer major complications, markedly lower rates of DGE, and shorter hospitalization. Stratified analyses supported enucleation even for tumors close to the MPD and identified 2.5 cm as the size threshold for greatest complication benefit. Long-term follow-up demonstrated superior quality of life, lower new-onset diabetes, and higher return-to-work rates after enucleation. Prospective studies with systematic MPD distance assessment and longitudinal patient-reported outcomes are warranted to further validate these findings.
eMethods. Study design, inclusion and exclusion criteria, surgical procedures, statistical analysis, and invasiveness-grade classification
eResults. Detailed perioperative outcomes, pancreatic fistula management, quality-of-life and functional outcomes, and oncologic outcomes by invasiveness grade
eDiscussion. Generalizability of length of stay and stratified oncologic outcomes
eTable 1. Baseline covariate balance before propensity score matching (standardized mean differences)
eTable 2. Baseline covariate balance after propensity score matching (standardized mean differences)
eTable 3. Doubly robust sensitivity analysis adjusting for residual tumor size imbalance after propensity score matching
eTable 4. Distribution of surgical approaches by treatment group (before PSM)
eTable 5. Perioperative outcomes in the propensity score-matched cohort (217 pairs)
eTable 6. Complete POPF grade distribution by ISGPS 2016 classification, before and after propensity score matching
eTable 7. Detailed breakdown of complications by Clavien-Dindo grade, ISGPS PPH grade, and ISGPS DGE grade
eTable 8. Multivariable logistic regression for factors independently associated with clinically relevant POPF, Clavien-Dindo >= III
eTable 9. Multivariable logistic regression incorporating tumor-to-main pancreatic duct distance as a continuous covariate, with interaction analysis
eTable 10. Complete quality of life outcomes (EORTC QLQ-C30 and QLQ-PAN26), endocrine/exocrine function, nutritional status, and social recovery (n = 551)
eTable 11. Quality of life outcomes in the propensity score-matched cohort (sensitivity analysis)
eTable 12. Worst-case sensitivity analysis for quality of life outcomes
eTable 13. Pre-specified subgroup analysis by tumor size and pathological type
eTable 14. Tumor-type specific propensity score matching analysis
eTable 15. Surgical approach subgroup analysis: robotic enucleation vs open enucleation, and robotic enucleation vs robotic PD
eTable 16. Oncologic outcomes by pathological invasiveness grade and surgical procedure
eFigure 1. Patient selection flow diagram
eFigure 2. Pathological tumor type distribution
eFigure 3. Covariate balance before and after propensity score matching (love plot)
eFigure 4. Distribution of postoperative pancreatic fistula grades by ISGPS classification
eFigure 5. Postoperative outcomes after propensity score matching (217 pairs)
eFigure 6. Multivariable logistic regression forest plot
eFigure 7. Outcomes stratified by tumor-to-main pancreatic duct distance
eFigure 8. Detailed quality of life comparison
eFigure 9. Subgroup analysis by tumor size and pathological type
eFigure 10. IPMN-specific subgroup analysis
eFigure 11. Tumor-type specific propensity score matching
eFigure 12. Robotic versus laparoscopic platform compare
Data Sharing Statement
References
- 1.Dasari A, Shen C, Halperin D, et al. Trends in the incidence, prevalence, and survival outcomes in patients with neuroendocrine tumors in the United States. JAMA Oncol. 2017;3(10):1335-1342. doi: 10.1001/jamaoncol.2017.0589 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.van Huijgevoort NCM, Del Chiaro M, Wolfgang CL, van Hooft JE, Besselink MG. Diagnosis and management of pancreatic cystic neoplasms: current evidence and guidelines. Nat Rev Gastroenterol Hepatol. 2019;16(11):676-689. doi: 10.1038/s41575-019-0195-x [DOI] [PubMed] [Google Scholar]
- 3.European Study Group on Cystic Tumours of the Pancreas . European evidence-based guidelines on pancreatic cystic neoplasms. Gut. 2018;67(5):789-804. doi: 10.1136/gutjnl-2018-316027 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Liu Q, Dai M, Guo J, et al. Long-term survival, quality of life, and molecular features of the patients with solid pseudopapillary neoplasm of the pancreas: a retrospective case study of 454 cases. Ann Surg. 2023;278(6):1009-1017. doi: 10.1097/SLA.0000000000005842 [DOI] [PubMed] [Google Scholar]
- 5.Das S, Dasari A. Epidemiology, incidence, and prevalence of neuroendocrine neoplasms: are there global differences? Curr Oncol Rep. 2021;23(4):43. doi: 10.1007/s11912-021-01029-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Allam M, Hidalgo Salinas C, Machairas N, Kostakis ID, Watkins J, Fusai GK. Solid pseudopapillary neoplasms of the pancreas: a single-center experience and review of the literature. J Gastrointest Cancer. 2022;53(2):460-465. doi: 10.1007/s12029-021-00638-6 [DOI] [PubMed] [Google Scholar]
- 7.Tanaka M, Fernández-del Castillo C, Adsay V, et al. ; International Association of Pancreatology . International consensus guidelines 2012 for the management of IPMN and MCN of the pancreas. Pancreatology. 2012;12(3):183-197. doi: 10.1016/j.pan.2012.04.004 [DOI] [PubMed] [Google Scholar]
- 8.Tanaka M, Fernández-Del Castillo C, Kamisawa T, et al. Revisions of international consensus Fukuoka guidelines for the management of IPMN of the pancreas. Pancreatology. 2017;17(5):738-753. doi: 10.1016/j.pan.2017.07.007 [DOI] [PubMed] [Google Scholar]
- 9.Inchauste SM, Lanier BJ, Libutti SK, et al. Rate of clinically significant postoperative pancreatic fistula in pancreatic neuroendocrine tumors. World J Surg. 2012;36(7):1517-1526. doi: 10.1007/s00268-012-1598-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bassi C, Marchegiani G, Dervenis C, et al. ; International Study Group on Pancreatic Surgery (ISGPS) . The 2016 update of the International Study Group (ISGPS) definition and grading of postoperative pancreatic fistula: 11 years after. Surgery. 2017;161(3):584-591. doi: 10.1016/j.surg.2016.11.014 [DOI] [PubMed] [Google Scholar]
- 11.Pedrazzoli S. Surgical treatment of pancreatic cancer: Currently debated topics on morbidity, mortality, and lymphadenectomy. Surg Oncol. 2022;45:101858. doi: 10.1016/j.suronc.2022.101858 [DOI] [PubMed] [Google Scholar]
- 12.Shen X, Yang X. Comparison of outcomes of enucleation vs. standard surgical resection for pancreatic neoplasms: a systematic review and meta-analysis. Front Surg. 2022;8:744316. doi: 10.3389/fsurg.2021.744316 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Heidsma CM, Tsilimigras DI, van Dieren S, et al. Indications and outcomes of enucleation versus formal pancreatectomy for pancreatic neuroendocrine tumors. HPB (Oxford). 2021;23(3):413-421. doi: 10.1016/j.hpb.2020.06.015 [DOI] [PubMed] [Google Scholar]
- 14.Crippa S, Bassi C, Salvia R, Falconi M, Butturini G, Pederzoli P. Enucleation of pancreatic neoplasms. Br J Surg. 2007;94(10):1254-1259. doi: 10.1002/bjs.5833 [DOI] [PubMed] [Google Scholar]
- 15.Cauley CE, Pitt HA, Ziegler KM, et al. Pancreatic enucleation: improved outcomes compared to resection. J Gastrointest Surg. 2012;16(7):1347-1353. doi: 10.1007/s11605-012-1893-7 [DOI] [PubMed] [Google Scholar]
- 16.Beger HG. Benign tumors of the pancreas—radical surgery versus parenchyma-sparing local resection—the challenge facing surgeons. J Gastrointest Surg. 2018;22(3):562-566. doi: 10.1007/s11605-017-3644-2 [DOI] [PubMed] [Google Scholar]
- 17.Beger HG, Mayer B, Poch B. Duodenum-preserving pancreatic head resection in 1063 patients for benign, premalignant cystic, and neuroendocrine neoplasms - short-term surgical outcomes and risk of recurrence - results of a systematic review. HPB (Oxford). 2025;27(12):1489-1502. doi: 10.1016/j.hpb.2025.09.002 [DOI] [PubMed] [Google Scholar]
- 18.Giuliani T, De Pastena M, Paiella S, et al. Pancreatic enucleation patients share the same quality of life as the general population at long-term follow-up: a propensity score-matched analysis. Ann Surg. 2023;277(3):e609-e616. doi: 10.1097/SLA.0000000000004911 [DOI] [PubMed] [Google Scholar]
- 19.Li Z, Shi Y, Zhuo Q, et al. Main pancreatic duct exposure, repair, or reconstruction during minimally invasive pancreatic enucleation: a long-term metabolic outcome from a prospective cohort study (CSPAC-MIEN-1). Int J Surg. 2025;111(11):8001-8014. doi: 10.1097/JS9.0000000000003016 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Zhou Y, Zhao M, Wu L, Ye F, Si X. Short- and long-term outcomes after enucleation of pancreatic tumors: an evidence-based assessment. Pancreatology. 2016;16(6):1092-1098. doi: 10.1016/j.pan.2016.07.006 [DOI] [PubMed] [Google Scholar]
- 21.Kaiser J, Alhalabi KT, Hinz U, et al. Enucleation for low-grade branch duct intraductal papillary mucinous neoplasms: Long-term follow-up. Surgery. 2022;172(3):968-974. doi: 10.1016/j.surg.2022.04.035 [DOI] [PubMed] [Google Scholar]
- 22.Beger HG, Mayer B, Poch B. Long-term oncologic outcome following duodenum-preserving pancreatic head resection for benign tumors, cystic neoplasms, and neuroendocrine tumors: systematic review and meta-analysis. Ann Surg Oncol. 2024;31(7):4637-4653. doi: 10.1245/s10434-024-15222-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Wolk S, Distler M, Kersting S, Weitz J, Saeger HD, Grützmann R. Evaluation of central pancreatectomy and pancreatic enucleation as pancreatic resections–a comparison. Int J Surg. 2015;22:118-124. doi: 10.1016/j.ijsu.2015.07.712 [DOI] [PubMed] [Google Scholar]
- 24.Duconseil P, Marchese U, Ewald J, et al. A pancreatic zone at higher risk of fistula after enucleation. World J Surg Oncol. 2018;16(1):177. doi: 10.1186/s12957-018-1476-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Turrini O, Schmidt CM, Pitt HA, et al. Side-branch intraductal papillary mucinous neoplasms of the pancreatic head/uncinate: resection or enucleation? HPB (Oxford). 2011;13(2):126-131. doi: 10.1111/j.1477-2574.2010.00256.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Asano Y, Kato H, Arakawa S, et al. Clinical outcomes of organ-preserving pancreatectomy for benign or low-grade malignant pancreatic tumors: a multicenter nationwide survey in Japan. J Hepatobiliary Pancreat Sci. 2022;29(8):898-910. doi: 10.1002/jhbp.1154 [DOI] [PubMed] [Google Scholar]
- 27.Paiella S, De Pastena M, Faustini F, et al. Central pancreatectomy for benign or low-grade malignant pancreatic lesions - a single-center retrospective analysis of 116 cases. Eur J Surg Oncol. 2019;45(5):788-792. doi: 10.1016/j.ejso.2018.11.021 [DOI] [PubMed] [Google Scholar]
- 28.Zu G, Chen W, Wu D, Zhang Y, Chen X. Clinical outcomes of minimally invasive duodenum-preserving pancreatic head resection. BMC Surg. 2023;23(1):288. doi: 10.1186/s12893-023-02170-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.de Graaf N, Zwart MJW, van Hilst J, et al. Early experience with robotic pancreatoduodenectomy versus open pancreatoduodenectomy: nationwide propensity-score-matched analysis. Br J Surg. 2024;111(2):znae043. doi: 10.1093/bjs/znae043 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Austin PC. Optimal caliper widths for propensity-score matching when estimating differences in means and differences in proportions in observational studies. Pharm Stat. 2011;10(2):150-161. doi: 10.1002/pst.433 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative . The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61(4):344-349. doi: 10.1016/j.jclinepi.2007.11.008 [DOI] [PubMed] [Google Scholar]
- 32.Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004;240(2):205-213. doi: 10.1097/01.sla.0000133083.54934.ae [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Clavien PA, Barkun J, de Oliveira ML, et al. The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg. 2009;250(2):187-196. doi: 10.1097/SLA.0b013e3181b13ca2 [DOI] [PubMed] [Google Scholar]
- 34.Wente MN, Veit JA, Bassi C, et al. Postpancreatectomy hemorrhage (PPH): an International Study Group of Pancreatic Surgery (ISGPS) definition. Surgery. 2007;142(1):20-25. doi: 10.1016/j.surg.2007.02.001 [DOI] [PubMed] [Google Scholar]
- 35.Wente MN, Bassi C, Dervenis C, et al. Delayed gastric emptying (DGE) after pancreatic surgery: a suggested definition by the International Study Group of Pancreatic Surgery (ISGPS). Surgery. 2007;142(5):761-768. doi: 10.1016/j.surg.2007.05.005 [DOI] [PubMed] [Google Scholar]
- 36.Ou H, Chen M, Qin K, et al. Short-term and long-term outcomes of robotic enucleation of tumors located in the pancreatic head and uncinate process. Ann Surg. 2025;282(1):86-92. doi: 10.1097/SLA.0000000000006198 [DOI] [PubMed] [Google Scholar]
- 37.Bursac Z, Gauss CH, Williams DK, Hosmer DW. Purposeful selection of variables in logistic regression. Source Code Biol Med. 2008;3(1):17. doi: 10.1186/1751-0473-3-17 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.O’Brien RM. A caution regarding rules of thumb for variance inflation factors. Qual Quant. 2007;41(5):673-690. doi: 10.1007/s11135-006-9018-6 [DOI] [Google Scholar]
- 39.Harrell FE Jr, Lee KL, Mark DB. Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med. 1996;15(4):361-387. doi: 10.1002/(SICI)1097-0258(19960229)15:4<361::AID-SIM168>3.0.CO;2-4 [DOI] [PubMed] [Google Scholar]
- 40.Li G, Taljaard M, Van Den Heuvel ER, et al. An introduction to multiplicity issues in clinical trials: the what, why, when and how. Int J Epidemiol. Published online December 26, 2016 doi: 10.1093/ije/dyw320 [DOI] [PubMed] [Google Scholar]
- 41.Aaronson NK, Ahmedzai S, Bergman B, et al. The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst. 1993;85(5):365-376. doi: 10.1093/jnci/85.5.365 [DOI] [PubMed] [Google Scholar]
- 42.Fitzsimmons D, Johnson CD, George S, et al. ; EORTC Study Group on Quality of Life . Development of a disease specific quality of life (QoL) questionnaire module to supplement the EORTC core cancer QoL questionnaire, the QLQ-C30 in patients with pancreatic cancer. Eur J Cancer. 1999;35(6):939-941. doi: 10.1016/S0959-8049(99)00047-7 [DOI] [PubMed] [Google Scholar]
- 43.White IR, Carlin JB. Bias and efficiency of multiple imputation compared with complete-case analysis for missing covariate values. Stat Med. 2010;29(28):2920-2931. doi: 10.1002/sim.3944 [DOI] [PubMed] [Google Scholar]
- 44.Jilesen APJ, van Eijck CHJ, Busch ORC, van Gulik TM, Gouma DJ, van Dijkum EJMN. Postoperative outcomes of enucleation and standard resections in patients with a pancreatic neuroendocrine tumor. World J Surg. 2016;40(3):715-728. doi: 10.1007/s00268-015-3341-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Brient C, Regenet N, Sulpice L, et al. Risk factors for postoperative pancreatic fistulization subsequent to enucleation. J Gastrointest Surg. 2012;16(10):1883-1887. doi: 10.1007/s11605-012-1971-x [DOI] [PubMed] [Google Scholar]
- 46.Tang W, Gao Y, Zhao Z, Kang Y, Xu Y, An L. Intraoperative ultrasound guided robotic pancreatic enucleation, does a distance of 3 mm still matter? Surg Innov. 2023;30(2):166-175. doi: 10.1177/15533506221092496 [DOI] [PubMed] [Google Scholar]
- 47.Callery MP, Pratt WB, Kent TS, Chaikof EL, Vollmer CM Jr. A prospectively validated clinical risk score accurately predicts pancreatic fistula after pancreatoduodenectomy. J Am Coll Surg. 2013;216(1):1-14. doi: 10.1016/j.jamcollsurg.2012.09.002 [DOI] [PubMed] [Google Scholar]
- 48.Di Martino M, de la Hoz Rodriguez Á, Saibanti A, et al. Pancreatic exocrine insufficiency after pancreatic resection: a systematic review. BMC Surg. 2025;25(1):53. doi: 10.1186/s12893-025-02787-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Thomas AS, Huang Y, Kwon W, et al. Prevalence and risk factors for pancreatic insufficiency after partial pancreatectomy. J Gastrointest Surg. 2022;26(7):1425-1435. doi: 10.1007/s11605-022-05302-3 [DOI] [PubMed] [Google Scholar]
- 50.Partelli S, Gaujoux S, Boninsegna L, et al. Pattern and clinical predictors of lymph node involvement in nonfunctioning pancreatic neuroendocrine tumors (NF-PanNETs). JAMA Surg. 2013;148(10):932-939. doi: 10.1001/jamasurg.2013.3376 [DOI] [PubMed] [Google Scholar]
- 51.Liu R, Liu Q, Wang Z. Expert consensus on pancreatic duct (repair) surgery. Int Surg. 2026;9:7-13. doi: 10.1016/j.isurg.2025.12.002 [DOI] [Google Scholar]
- 52.Liu R, Abu Hilal M, Besselink MG, et al. International consensus guidelines on robotic pancreatic surgery in 2023. Hepatobiliary Surg Nutr. 2024;13(1):89-104. doi: 10.21037/hbsn-23-132 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Liu R, Liu Q. Pancreatic duct surgery: from innovation to implementation. Int Surg. 2023;6:8-11. doi: 10.1016/j.isurg.2022.11.001 [DOI] [Google Scholar]
- 54.Bolm L, Nebbia M, Wei AC, et al. Long-term outcomes of parenchyma-sparing and oncologic resections in patients with nonfunctional pancreatic neuroendocrine tumors <3 cm (PANDA). Ann Surg. 2022;276(3):522-531. doi: 10.1097/SLA.0000000000005559 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Karam E, Nassar A, Elkurdi S, et al. Enucleation for sporadic nonfunctioning pancreatic neuroendocrine tumors larger than 2 centimeters is associated with equivalent morbidity and survival compared to smaller tumors: a multi-institutional study. Neuroendocrinology. 2024;114(11):1034-1044. doi: 10.1159/000541078 [DOI] [PubMed] [Google Scholar]
- 56.Edwin B, Sahakyan MA, Abu Hilal M, et al. ; EAES Consensus Conference Study Group . Laparoscopic surgery for pancreatic neoplasms: the European association for endoscopic surgery clinical consensus conference. Surg Endosc. 2017;31(5):2023-2041. doi: 10.1007/s00464-017-5414-3 [DOI] [PubMed] [Google Scholar]
- 57.Guerra F, Giuliani G, Bencini L, Bianchi PP, Coratti A. Minimally invasive versus open pancreatic enucleation: aystematic review and meta-analysis of surgical outcomes. J Surg Oncol. 2018;117(7):1509-1516. doi: 10.1002/jso.25026 [DOI] [PubMed] [Google Scholar]
- 58.Zhao G, Liu Q, Zhao Z, et al. The standardized technique and surgical video of robotic pancreaticoduodenectomy at the Chinese PLA General Hospital. Updates Surg. 2022;74(1):245-254. doi: 10.1007/s13304-021-01149-6 [DOI] [PubMed] [Google Scholar]
- 59.de Graaf N, Augustinus S, Wellner UF, et al. ; Global Audits on Pancreatic Surgery Group (GAPASURG) . Transatlantic differences in the use and outcome of minimally invasive pancreatoduodenectomy: an international multi-registry analysis. Surg Endosc. 2024;38(12):7099-7111. doi: 10.1007/s00464-024-11161-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eMethods. Study design, inclusion and exclusion criteria, surgical procedures, statistical analysis, and invasiveness-grade classification
eResults. Detailed perioperative outcomes, pancreatic fistula management, quality-of-life and functional outcomes, and oncologic outcomes by invasiveness grade
eDiscussion. Generalizability of length of stay and stratified oncologic outcomes
eTable 1. Baseline covariate balance before propensity score matching (standardized mean differences)
eTable 2. Baseline covariate balance after propensity score matching (standardized mean differences)
eTable 3. Doubly robust sensitivity analysis adjusting for residual tumor size imbalance after propensity score matching
eTable 4. Distribution of surgical approaches by treatment group (before PSM)
eTable 5. Perioperative outcomes in the propensity score-matched cohort (217 pairs)
eTable 6. Complete POPF grade distribution by ISGPS 2016 classification, before and after propensity score matching
eTable 7. Detailed breakdown of complications by Clavien-Dindo grade, ISGPS PPH grade, and ISGPS DGE grade
eTable 8. Multivariable logistic regression for factors independently associated with clinically relevant POPF, Clavien-Dindo >= III
eTable 9. Multivariable logistic regression incorporating tumor-to-main pancreatic duct distance as a continuous covariate, with interaction analysis
eTable 10. Complete quality of life outcomes (EORTC QLQ-C30 and QLQ-PAN26), endocrine/exocrine function, nutritional status, and social recovery (n = 551)
eTable 11. Quality of life outcomes in the propensity score-matched cohort (sensitivity analysis)
eTable 12. Worst-case sensitivity analysis for quality of life outcomes
eTable 13. Pre-specified subgroup analysis by tumor size and pathological type
eTable 14. Tumor-type specific propensity score matching analysis
eTable 15. Surgical approach subgroup analysis: robotic enucleation vs open enucleation, and robotic enucleation vs robotic PD
eTable 16. Oncologic outcomes by pathological invasiveness grade and surgical procedure
eFigure 1. Patient selection flow diagram
eFigure 2. Pathological tumor type distribution
eFigure 3. Covariate balance before and after propensity score matching (love plot)
eFigure 4. Distribution of postoperative pancreatic fistula grades by ISGPS classification
eFigure 5. Postoperative outcomes after propensity score matching (217 pairs)
eFigure 6. Multivariable logistic regression forest plot
eFigure 7. Outcomes stratified by tumor-to-main pancreatic duct distance
eFigure 8. Detailed quality of life comparison
eFigure 9. Subgroup analysis by tumor size and pathological type
eFigure 10. IPMN-specific subgroup analysis
eFigure 11. Tumor-type specific propensity score matching
eFigure 12. Robotic versus laparoscopic platform compare
Data Sharing Statement
