Abstract
Background/Aim: Pancreaticoduodenectomy (PD) is the standard curative option for pancreatic head carcinoma, but R1 resections remain frequent. The superior mesenteric artery (SMA) “artery-first” approach (AFA) may improve oncologic clearance and operative safety. We conducted a systematic review and meta-analysis comparing AFA versus standard PD.
Materials and Methods: Following the PRISMA 2020 guidelines, PubMed and Scopus were searched to January 27, 2025. Comparative studies reporting R0 resection or survival outcomes were eligible. Data were extracted independently, and random-effects meta-analyses were performed for dichotomous (risk ratios, RR) and continuous outcomes (mean differences, MD). Sensitivity analyses included fixed-effect models, alternative continuity corrections, and exclusion of zero-event studies. Publication bias was assessed with funnel plots and Egger’s test.
Results: Twenty-two studies (2 randomized trials, 20 retrospective; 945 AFA, 1,173 standard PD patients) were included. R0 resection was reported in 15 studies. Pooled analysis favored AFA (RR: 1.13, 95% CI=1.08-1.19; I2=9.1%), robust across sensitivity analyses, with no evidence of publication bias (Egger p=0.47). Estimated blood loss was significantly reduced with AFA (MD −86.7 ml, 95% CI=−164.8 to −8.7; I2=63.8%), though results were moderately sensitive to individual studies. Other perioperative outcomes – including operative time, complications, and hospital stay – showed no consistent differences. Long-term survival outcomes were inconsistently reported but suggested at least equivalence. Subgroup analysis indicated that randomized trials did not consistently replicate the R0 advantage seen in retrospective series.
Conclusion: AFA for PD is associated with higher R0 resection rates and lower blood loss compared to standard PD, without added morbidity. Observational data strongly support AFA, but confirmatory evidence from well-powered randomized trials remains essential.
- Artery-first approach
- superior mesenteric artery (SMA)
- pancreaticoduodenectomy
- pancreatic neoplasms
- R0 resection
- meta-analysis
- review
Introduction
Pancreatic ductal adenocarcinoma (PDAC) arising from the head of the pancreas remains one of the most aggressive solid malignancies, characterized by late presentation, limited therapeutic options, and dismal prognosis (1). Globally, pancreatic cancer is responsible for over 331,000 deaths annually and ranks as the seventh leading cause of cancer-related mortality, with a 5-year survival rate that rarely exceeds 5% (2, 3). Despite improvements in chemotherapy and supportive care, surgical resection remains the only potentially curative treatment, with pancreatoduodenectomy (PD) serving as the gold standard for resectable tumors in the pancreatic head and periampullary region (4).
Complete oncological resection with negative margins (R0 resection) is the single most important determinant of survival (5). However, standardized pathologic examinations often reveal microscopic margin involvement (R1 resection) in more than 50% of cases, especially at the posterior and medial margins adjacent to the superior mesenteric artery (SMA) (6, 7). To overcome this, the “artery-first” approach (AFA) has been proposed, involving early exposure and dissection of the SMA before transection of the pancreas and other structures. The rationale of AFA is to facilitate assessment of resectability, reduce intraoperative blood loss, improve mesopancreas clearance, and enhance the likelihood of achieving an R0 resection (10).
Several variations of AFA have been described, including the posterior, medial uncinate, left posterior, right posterior, inferior infracolic, and supracolic anterior approaches (Table I). However, the true oncological and perioperative benefits of AFA compared to standard PD remain controversial. While some studies suggest superior oncological outcomes and improved survival with AFA, others have found no significant difference. The aim of this systematic review was to critically evaluate the available evidence comparing the artery-first approach with standard PD in terms of surgical, oncological, and survival outcomes.
Summary of the various techniques of SMA-first approach.
Materials and Methods
This review was conducted according to the PRISMA 2020 guidelines (11). A systematic literature search was performed in PubMed and Scopus up to January 27, 2025. Search terms included: (“Whipple” OR “Whipple surgery” OR “pancreatoduodenectomy” OR “pancreaticoduodenectomy” OR “duodenopancreatectomy”) AND (“superior mesenteric artery first” OR “SMA-first” OR “artery first” OR “mesenteric first”).
Studies were included if they: (i) were published in English, (ii) reported original data comparing SMA-first PD with standard PD, and (iii) provided outcomes on overall survival, disease-free survival, or R0 resection rates. Exclusion criteria included reviews, editorials, animal studies, and studies without direct comparison.
Two investigators independently screened titles (MS and NG) and abstracts using Rayyan (12), followed by full-text evaluation. Discrepancies were resolved by consensus with a third reviewer (MK). Data extraction was performed by three investigators and included study design, country, number of patients, demographics, perioperative outcomes (blood loss, operative time, morbidity, mortality), oncological outcomes (R0 resection, lymph node yield), and survival outcomes (1-, 2-, 3-, and 5-year OS and DFS). The primary outcomes were overall survival and disease-free survival, while secondary outcomes included perioperative mortality, blood loss, transfusion requirements, portal vein or SMV resection, postoperative complications, hospital stay, and recurrence.
Statistical analysis and meta-analysis were performed using inverse-variance random-effects models with DerSimonian–Laird estimation of between-study variance. For dichotomous outcomes (e.g., R0 resection), we pooled risk ratios (RR) using log-transformed RRs and applied a continuity correction of 0.5 for studies with zero events in one arm. For continuous outcomes (e.g., estimated blood loss), mean differences (MD) were pooled using inverse-variance weighting. Between-study heterogeneity was quantified by Cochran’s Q and the between-study variance tau-squared (τ2). Publication bias for outcomes with sufficient studies was assessed by funnel plot visualization and Egger’s regression test (regression of the standardized effect on precision). All analyses were implemented in Python (NumPy/SciPy/Matplotlib) within the study workspace. Two-sided p-values <0.05 were considered statistically significant.
Results
The initial search yielded 555 articles, of which 305 were selected for abstract review. Following screening, 59 articles underwent full-text review, and 22 studies met the inclusion criteria (10, 13-33). Three additional articles were excluded due to lack of full text despite author correspondence. No additional studies were identified through reference searching. A PRISMA flow diagram is shown in Figure 1.
Prisma Flow Diagram. Flowchart summarizing the identification, screening, eligibility, and inclusion of studies comparing the AFA versus standard PD. A total of 555 records were identified through database searching (PubMed and Scopus). After removal of duplicates and screening of titles and abstracts, 59 full-text articles were assessed for eligibility. Twenty-two studies met inclusion criteria and were included in the quantitative synthesis (meta-analysis). Three full-text articles were unavailable. AFA: Artery-first approach; PD: pancreaticoduodenectomy.
The included studies comprised two randomized controlled trials and twenty retrospective studies, published between 2009 and 2024. These studies were conducted across centers in China, Japan, India, Australia, Egypt, Poland, Romania, and Spain. In total, the SMA-first cohort included 945 patients, while the standard PD cohort included 1,173 patients. Patient age ranged between 54.5 and 72 years for the SMA-first group and 60 to 73 years for the standard PD group.
With respect to survival, one-year overall survival ranged from 19.9% to 92.8% in the SMA-first group and from 20.8% to 82.9% in the standard PD group. Two-year overall survival ranged from 49.7% to 84.4% in the SMA-first group compared to 38.6% to 69.3% in the standard PD group. Three-year overall survival was reported as 21% to 84.4% for SMA-first and 12% to 64% for standard PD. Only one study reported 5-year overall survival, showing a significant advantage for SMA-first (40.3% vs. 5.9%). Disease-free survival was less consistently reported. One-year DFS ranged between 47% and 92.9% in the SMA-first group and 62% to 71.9% in the standard group. Two- and three-year DFS were available in limited studies, but consistently favored the SMA-first approach, while no studies provided 5-year DFS data.
Oncological outcomes demonstrated variable but generally favorable trends with the artery-first approach. R0 resection rates ranged from 42.9% to 100% in SMA-first resections and 57.1% to 100% in standard PD. Lymph node yield was also higher in some SMA-first series, reaching up to 40 nodes compared to 35 nodes in the standard group. Recurrence data were limited but suggested delayed recurrence in SMA-first patients.
Perioperative outcomes showed comparable safety between the two techniques. Operative time ranged from 255 to 564 minutes for SMA-first and 264 to 540 minutes for standard PD. Estimated blood loss ranged from 100 to 1,307 ml for SMA-first and 175 to 1,568 ml for standard PD, with some studies reporting lower transfusion rates with AFA. Hospital stay varied widely but was not significantly different. Thirty-day mortality ranged from 2.2% to 6.6% for SMA-first and from 1.2% to 15.6% for standard PD (Table II).
Oncological and perioperative outcomes of SMA-first approach.
Meta-analysis of R0 resection rates included 15 studies with a total of 619 patients in the SMA-first group and 744 patients in the standard PD group and demonstrated a pooled risk ratio of 1.13 (95% CI=1.08-1.19) favoring the artery-first (SMA-first) approach over standard pancreatoduodenectomy (DerSimonian–Laird random-effects model; τ2=0.0008; Q=15.40, p=0.351, Figure 2). Heterogeneity of the R0 outcome was low. A funnel plot for R0 resection was visually inspected and Egger’s regression test did not indicate small-study effects (intercept=0.337, SE=0.450, t=0.748, p=0.468, Figure 3).
Forest plot of R0 resection comparing artery-first (SMA-first) versus standard PD. Individual study risk ratios (black points) and 95% confidence intervals (gray lines) are plotted on a logarithmic x-axis; the pooled random-effects estimate is shown as a blue square. Vertical dashed red line indicates no effect (RR=1). Studies with zero events were continuity-corrected (0.5) for pooling. The pooled risk ratio is RR=1.13 (95% CI=1.08-1.19), k=15 studies (DerSimonian–Laird random effects). PD: Pancreatoduodenectomy; SMA: superior mesenteric artery; CI: confidence interval; RR: relative risk.
Egger test results (regression of standard normal deviate on precision; k=15 studies): the intercept=0.337, SE (intercept)=0.450, t=0.748, p=0.468. This indicates no evidence of small-study/publication bias by Egger’s test (p=0.47), though interpretation is limited. SE: Standard error.
Meta-analysis of estimated blood loss (EBL) included seven studies and showed a pooled mean difference of −86.7 ml (95% CI=−164.8 to −8.7 ml), favoring the artery-first approach (DerSimonian–Laird random-effects model). Substantial heterogeneity was present for EBL (τ2≈6453; Q=16.59, p=0.011, Figure 4), suggesting notable between-study variability in blood loss reporting and surgical practices; hence, these pooled EBL results should be interpreted cautiously.
Forest plot of EBL comparing artery-first (SMA-first) versus standard PD. Individual study mean differences (black points) and 95% CIs (gray lines) are plotted with a vertical dashed red line at zero mean difference; negative values favor SMA-first (less blood loss). The pooled mean difference is MD=−86.7 ml (95% CI=−164.8 to −8.7 ml), k=7 studies (DerSimonian–Laird random effects). Note: between-study heterogeneity for EBL was substantial. EBL: Estimated blood loss; SMA: superior mesenteric artery, MD: mean difference; CI: confidence interval; PD: pancreaticoduodenectomy.
Discussion
This systematic review demonstrates that the artery-first approach is a feasible and safe alternative to standard PD, with potential oncological advantages. Several studies indicated higher R0 resection rates and improved medium-term survival with AFA (10, 13-32). The theoretical benefits of early SMA exposure – better resectability assessment, improved mesopancreas clearance, and reduced blood loss – are supported by clinical data, although definitive proof remains limited.
However, the evidence is marked by heterogeneity. The majority of studies were retrospective, with significant variation in patient selection, definitions of margin status, and follow-up duration. Only two randomized controlled trials have been conducted, both with limited power. Furthermore, definitions of R0 resection remain inconsistent, with some using the 1-mm clearance rule while others only reporting direct involvement. This variability complicates pooled interpretation of oncological outcomes.
The six core artery-first techniques for pancreatoduodenectomy all focus on early exposure of the SMA prior to irreversible operative steps, such as division of the pancreatic neck (Figure 5). This early dissection facilitates assessment of resectability and may enhance oncological clearance. The posterior approach gains access to the SMA from the retroperitoneum, typically after a Cattell–Braasch maneuver, allowing posterior clearance before division of the pancreas. The right, or medial uncinate, approach proceeds retrogradely from the uncinate process toward the SMA, with early ligation of the inferior pancreaticoduodenal artery, thereby minimizing blood loss and aiding oncological dissection. The mesenteric, or inferior infracolic, approach begins in the infracolic region medial to the duodenojejunal flexure and proceeds along the mesentery to expose the SMA at an early stage of the operation. The left-posterior approach involves dissection of the SMA in a clockwise fashion from the left side, with leftward mobilization of the jejunum. This approach is especially valuable for identifying an aberrant right hepatic artery and improving margin assessment. The superior approach provides access through the lesser sac and hepatoduodenal ligament and is particularly advantageous in cases of tumors involving the common hepatic artery or the upper pancreatic head. Finally, the anterior, or inferior supracolic, approach enters the supracolic compartment and lesser sac. Although less commonly performed, it remains one of the recognized variations of the artery-first technique. Together, these six approaches represent technical adaptations of a unifying principle: establishing SMA control early to optimize both surgical safety and oncological outcomes. From a technical standpoint, AFA requires advanced surgical expertise and may involve a steep learning curve. In high-volume centers, it appears to be associated with less intraoperative blood loss and similar complication rates compared to standard PD. In lower-volume settings, however, its technical complexity could potentially increase perioperative risk.
The six core artery-first techniques. These approaches all center on early exposure of the SMA before irreversible steps (like pancreatic neck division), helping assess operability and potentially improving oncologic outcomes. SMA: Superior mesenteric artery.
The artery-first approach represents an important conceptual and technical evolution in pancreatic surgery. Its major strength lies in achieving early assessment of tumor–vessel relationships at the root of the mesentery, enabling safer dissection, improved vascular control, and avoidance of futile resections in cases of unsuspected arterial encasement. Furthermore, this approach facilitates a more systematic clearance of the mesopancreas – a recognized source of R1 involvement – thereby increasing the likelihood of achieving true R0 resection. However, the technique is not without limitations. It demands a deep understanding of complex peripancreatic vascular anatomy and meticulous dissection skills, limiting its reproducibility in lower-volume centers or among less-experienced surgeons. Heterogeneity across the six described artery-first variants and lack of standardization in nomenclature and operative steps further complicate comparisons between studies. Moreover, most published data derive from retrospective analyses with inherent selection bias and limited adjustment for tumor biology or neoadjuvant therapy. From an expert surgical perspective, the artery-first technique should not be viewed as a single operation but rather as a strategic framework that can be tailored to individual anatomy and tumor extent. When applied selectively by experienced hepatopancreatobiliary surgeons, it can optimize oncologic radicality and intraoperative decision-making. Nevertheless, widespread adoption should be preceded by structured training, standardized procedural definitions, and prospective evaluation. High-quality randomized trials integrating arterial-first strategies within modern multimodal treatment pathways will ultimately determine whether these technical refinements translate into meaningful survival benefit.
Limitations of this review include the predominance of retrospective data, heterogeneity in reported outcomes, and lack of standardized definitions across studies. Moreover, the paucity of long-term survival data restricts conclusions regarding the impact of AFA on cure rates. Nevertheless, the findings suggest that artery-first pancreatoduodenectomy may improve oncological clearance and medium-term survival in selected patients. Prospective, multicenter randomized controlled trials with standardized pathology protocols and adequate long-term follow-up are needed to determine whether AFA should become the standard of care.
Conclusion
The artery-first approach to pancreatoduodenectomy is safe, feasible, and may offer oncological advantages over the standard approach. Current evidence suggests higher R0 resection rates, lower intraoperative blood loss, and improved medium-term survival in selected patients. However, the evidence base remains heterogeneous and limited by methodological shortcomings. Further high-quality randomized controlled trials are required to define the role of AFA in modern pancreatic surgery.
Footnotes
Authors’ Contributions
Dimitrios Moris and Michail Skandalakis conceived and designed the study, performed data analysis, and drafted the manuscript. Nikolaos Gonidakis and Maria Koundouraki contributed to data extraction, verification, and statistical interpretation. Dimitrios Moris and Evangelos Mplevrakis supervised methodological accuracy, critically reviewed the manuscript, and provided expert input on surgical aspects of the artery-first technique. Sofia Theodoridou participated in literature screening, study selection, and figure preparation. All authors contributed to manuscript revision, read, and approved the final version of the article.
Conflicts of Interest
None to be declared.
Artificial Intelligence (AI) Disclosure
During the preparation of this manuscript, a large language model (ChatGPT, OpenAI) was used for language editing and stylistic improvements in select paragraphs. No sections involving the generation, analysis, or interpretation of research data were produced by generative AI. All scientific content was created and verified by the Authors. The Authors used ChatGPT (GPT-5, OpenAI, San Francisco, CA, USA) for the generation of Figure 5.
- Received October 7, 2025.
- Revision received October 18, 2025.
- Accepted October 29, 2025.
- Copyright © 2026 The Author(s). Published by the International Institute of Anticancer Research.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.











