Defining and predicting textbook outcomes in laparoscopic distal pancreatectomy

Abstract

BACKGROUND

Laparoscopic distal pancreatectomy (LDP) has emerged as the preferred approach for both benign and malignant lesions located in the pancreatic body and tail. Nevertheless, a notable deficiency persists in the absence of a standardized, procedure-specific metric for evaluating and comparing surgical quality. A composite measure termed “textbook outcome (TO)”, which encompasses key short-term endpoints, has been validated in laparoscopic pancreatoduodenectomy but has not yet been established in dedicated LDP cohorts. The definition and prediction of TO in this context could aid in facilitating cross-institutional benchmarking and fostering advancements in quality improvement.

AIM

To establish procedure-specific criteria for TO and identify independent predictors of TO failure in patients undergoing LDP.

METHODS

Consecutive patients who underwent LDP at a single high-volume pancreatic center between January 2015 and August 2022 were retrospectively analyzed. TO was defined as the absence of clinically relevant postoperative pancreatic fistula (grade B/C), post-pancreatectomy hemorrhage (grade B/C), severe complications (Clavien-Dindo ≥ III), readmission within 30 days, and in-hospital or 30-day mortality. Multivariable logistic regression was employed to identify independent predictors of TO failure, and a nomogram was constructed and internally validated.

RESULTS

Among 405 eligible patients, 286 (70.6%) attained TO. Multivariable analysis revealed that female sex [odds ratio (OR) = 0.62, 95% confidence interval (CI): 0.39-0.99] conferred a protective effect, while preoperative endoscopic ultrasound-guided fine-needle aspiration (OR = 2.66, 95%CI: 1.05-6.73), pancreatic portal hypertension (OR = 2.81, 95%CI: 1.06-7.45), and cystic-solid (OR = 2.51, 95%CI: 1.34-4.69) or solid lesions (OR = 1.91, 95%CI: 1.06-3.44) were independently associated with TO failure (all P < 0.05). The derived nomogram exhibited modest discrimination and calibration when assessed in both the training and validation datasets.

CONCLUSION

The proposed LDP-specific definition of TO is feasible and discriminative, and the developed nomogram provides an objective tool for individualized risk assessment.

Key Words: Laparoscopic distal pancreatectomy; Textbook outcome; Predictors; Risk prediction model; Nomogram

Core Tip: We established a laparoscopic distal pancreatectomy-specific textbook outcome composite absence of grade B/C postoperative pancreatic fistula, grade B/C post-pancreatectomy hemorrhage, Clavien-Dindo ≥ III morbidity, 30-day readmission, and in-hospital or 30-day mortality as a standardized quality indicator for distal pancreatic resection. Among 405 consecutive patients, textbook outcome was attained in 70.6%; female sex enhanced achievement, whereas pre-operative endoscopic ultrasound-guided fine-needle aspiration, pancreatic portal hypertension, and cystic-solid/solid pathology predicted failure. An internally validated nomogram is provided for individual pre-operative risk counselling.

INTRODUCTION

Minimally invasive surgery has progressed to the extent where laparoscopy has established itself as the standard approach for pancreatic interventions, including laparoscopic distal pancreatectomy (LDP)[1]. LDP presents numerous advantages over traditional open surgery, such as obviating the necessity for biliary-pancreatic-enteric reconstruction, simplifying operative procedures, and offering benefits such as reduced incision size, diminished blood loss, decreased occurrences of delayed gastric emptying, shorter hospitalization durations, and lower overall costs while upholding comparable long-term oncologic outcomes[2-4]. Consequently, LDP has evolved into a standard practice for both benign conditions (e.g., trauma, chronic pancreatitis, cystadenoma) and malignant lesions (e.g., ductal adenocarcinoma) affecting the pancreatic body and tail in high-volume centers worldwide[5,6].

Despite these advancements, LDP remains a technically demanding procedure with significant perioperative risks. The pancreas is situated deep within the retroperitoneum, encircled by critical structures such as the portal vein, superior mesenteric vessels, splenic vessels, stomach, colon, and spleen, which constrains the operative field[7]. The gland itself is delicate and prone to damage, with pancreatic juice possessing high corrosive properties, making even minor parenchymal injuries, incomplete duct closure, or vascular trauma potentially leading to severe complications like pancreatic fistula, hemorrhage, infection, or delayed gastric emptying[8,9]. Consequently, patient outcomes are heavily dependent on the surgeon’s laparoscopic expertise, anatomical knowledge, ability to manage emergencies, and team coordination. Therefore, meticulous and standardized perioperative care is essential to mitigate these risks.

While traditional quality metrics such as mortality, major morbidity, and length of stay (LOS) are easily measurable, they do not offer a comprehensive evaluation of perioperative quality[10,11]. Composite indicators such as “textbook outcome (TO)” are now considered as more effective than individual endpoints for assessing surgical quality. Initially introduced by Kolfschoten et al[12] in 2013 for colorectal surgery, TO represents an “all-or-none” composite measure[13] encompassing the absence of major complications, reoperation, readmission, prolonged hospitalization, and 30-day mortality. Since its inception, TO has been widely adopted in gastrointestinal, hepatobiliary, and pancreatic surgery as a sensitive discriminator between institutions and a practical tool for enhancing quality[14-17].

In 2020, the Dutch Pancreatic Cancer Group (DPCG) proposed the initial definition of TO for pancreatic surgery in a nationwide study involving 3341 pancreatoduodenectomies (PD) and distal pancreatectomies (DP)[18]. Subsequent researches have demonstrated that TO standardizes quality assessment, identifies modifiable determinants of an “ideal” clinical course, and reduces resource utilization[19,20]. Nonetheless, the majority of TO investigations in pancreatic surgery have predominantly concentrated on PD[21], with existing DP datasets frequently amalgamating open and laparoscopic cases[22], thereby creating a notable gap in evidence gap for LDP-specific TO.

This study aims to fill this research void by investigating the occurrence of TO in LDP, identifying its independent predictors, and constructing a visual risk prediction model using data from a single-center, prospectively maintained database. The results offer an evidence-based quality assessment tool for minimally invasive pancreatic surgery, advocate for the standardization of LDP procedures, expedite recovery, reduce healthcare expenses, and contribute Chinese data to future international guidelines. Through these efforts, we aspire to enhance the comprehension and implementation of TO in LDP, ultimately improving patient outcomes and surgical quality.

MATERIALS AND METHODS

Patients

This retrospective single-center study was carried out with the approval of the Ethics Committee of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology (No. TJ-IRB202407020). Sequential patients who underwent LDP between January 1, 2015, and August 31, 2022, were identified from our prospectively maintained pancreatic surgery database. Essential missing variables were obtained from the electronic medical records and the picture archiving and communication system.

Exclusion criteria comprised: (1) Cases of diagnostic laparoscopy or biopsy solely, or those converted to open surgery (n = 45); (2) Prior pancreatic operation requiring redo LDP (n = 8); and (3) Incomplete assessment of any TO component (n = 59). The final analysis included a cohort of 405 patients.

All procedures were performed by three senior surgical teams, each of which had surpassed the learning curve for LDP. Preoperative indications were assessed by a multidisciplinary team: Patients were considered eligible if they had benign or malignant lesions confined to the pancreatic body or tail, no involvement of the celiac axis or superior mesenteric artery, no severe comorbidities, and sufficient tolerance to pneumoperitoneum.

Definition

Pancreas-specific morbidities, namely post-pancreatectomy hemorrhage (PPH) and postoperative pancreatic fistula (POPF), were categorized according to the 2007 and 2016 criteria of the International Study Group of Pancreatic Surgery[23,24], respectively. Other complications were classified using Clavien-Dindo grading system[25], with a grade of III or higher considered severe complications.

In LDP, TO was defined as the simultaneous satisfaction of the following criteria: (1) Absence of POPF (grade B/C); (2) Absence of PPH (grade B/C); (3) Absence of severe complications (Clavien-Dindo ≥ III); (4) No readmission within 30 days after discharge; and (5) No in-hospital or 30-day mortality[18].

Statistical analyses

Statistical analyses were performed using R (version 4.4.1, R Core Team, 2024). A two-sided P < 0.05 was deemed statistically significant. The normality of continuous variables was assessed using the Shapiro-Wilk test. Normally distributed data are presented as mean ± SD and compared between groups using the independent-samples Student’s t-test. Non-normally distributed data are expressed as median (25^th, 75^th percentiles) and compared using the Mann-Whitney U test. Categorical variables are described as n (%) and compared using Pearson’s χ² test, continuity-corrected χ² test, or Fisher’s exact test, depending on expected cell frequencies. Univariable screening (P < 0.05) was followed by multivariable logistic regression, with backward elimination used to retain variables that minimized the Akaike information criterion. The dataset was randomly divided into a training set and a validation set in a 7:3 ratio. The risk-prediction model was developed using the training set and internally validated in the validation set. Discrimination was quantified using the concordance index (C-index), and calibration was evaluated using calibration plots.

RESULTS

Baseline characteristics

The baseline characteristics of the 405 patients who underwent LDP are summarized in Table 1. Patients who achieved TO were significantly younger, more often female, and exhibited lower body mass index and fewer cardiovascular comorbidity compared to the TO-failure group (all P < 0.05). Significant differences were also observed in lesion-related factors: The TO cohort had fewer cystic-solid and solid lesions, a lower incidence of pancreatic portal hypertension, decreased involvement of splenic vessels, lower rates of local invasion, and less frequent preoperative endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) utilization (all P < 0.05). Intraoperatively, cases meeting TO criteria displayed shorter anesthesia and operative durations, along with a higher rate of spleen preservation.

Table 1 Baseline characteristics of 405 patients undergoing laparoscopic distal pancreatectomy, n (%).

Variables	Total (n = 405)	TO (n = 286)	TO failure (n = 119)	P value
Age (years), median IQR	51.0 (39.0-60.0)	50.0 (37.0-58.0)	53.0 (42.0-61.0)	0.025a
Sex				0.004a
Female	245 (60.5)	186 (65)	59 (49.6)
Male	160 (39.5)	100 (35)	60 (50.4)
BMI, median IQR	22.5 (20.3-24.7)	22.3 (20.0-24.3)	23.2 (20.7-25.1)	0.045a
Smoker or ex-smoker	31 (7.7)	20 (7)	11 (9.2)	0.438
Diabetes	25 (6.2)	20 (7)	5 (4.2)	0.288
Cardiovascular disease	57 (14.1)	33 (11.5)	24 (20.2)	0.023a
Previous abdominal surgery	34 (8.4)	25 (8.7)	9 (7.6)	0.697
Asymptomatic	170 (42)	123 (43)	47 (39.5)	0.514
Weight loss	56 (13.8)	34 (11.9)	22 (18.5)	0.080
ASA score				0.425
I	64 (15.8)	49 (17.1)	15 (12.6)
II	279 (68.9)	196 (68.5)	83 (69.7)
III	62 (15.3)	41 (14.3)	21 (17.6)
Lesion size (cm), median IQR	3.5 (2.2-5.2)	3.4 (2.1-5.0)	3.7 (2.5-5.4)	0.063
Lesion location				0.514
Body	189 (46.7)	136 (47.6)	53 (44.5)
Neck	45 (11.1)	34 (11.9)	11 (9.2)
Tail	171 (42.2)	116 (40.6)	55 (46.2)
Pancreatic duct dilation	73 (18)	55 (19.2)	18 (15.1)	0.328
Lesion consistency				< 0.001a
Cystic	142 (35.1)	117 (40.9)	25 (21)
Cystic-solid	101 (24.9)	65 (22.7)	36 (30.3)
Solid	162 (40)	104 (36.4)	58 (48.7)
Pancreatitis	30 (7.4)	19 (6.6)	11 (9.2)	0.363
Pancreatic portal hypertension	27 (6.7)	13 (4.5)	14 (11.8)	0.008a
Splenic vessel involvement	56 (13.8)	33 (11.5)	23 (19.3)	0.039a
Preoperative EUS-FNA	21 (5.2)	10 (3.5)	11 (9.2)	0.017a
Pancreatic texture				0.746
Hard	58 (14.3)	42 (14.7)	16 (13.4)
Soft	347 (85.7)	244 (85.3)	103 (86.6)
Local invasion	18 (4.4)	8 (2.8)	10 (8.4)	0.013a
Anesthesia time (minute), median IQR	212.0 (180.0-255.0)	206.0 (174.0-245.0)	221.0 (191.0-285.5)	< 0.001a
Operation time (minute), median IQR	180.0 (150.0-225.0)	178.0 (142.0-210.0)	189.0 (160.0-245.5)	0.005a
Intraoperative blood loss (mL), median IQR	100.0 (50.0-100.0)	100.0 (50.0-100.0)	100.0 (50.0-175.0)	0.055
Intraoperative blood transfusion	16 (4)	10 (3.5)	6 (5)	0.467
Spleen preservation	179 (44.2)	139 (48.6)	40 (33.6)	0.006
Additional organ resection	4 (1)	2 (0.7)	2 (1.7)	0.363
Lymph node dissection	182 (44.9)	121 (42.3)	61 (51.3)	0.099
Malignancy	86 (21.2)	54 (18.9)	32 (26.9)	0.073
Histological diagnoses				0.033a
CP/PPC	34 (8.4)	22 (7.7)	12 (10.1)
Cyst	13 (3.2)	10 (3.5)	3 (2.5)
IPMN	20 (4.9)	18 (6.3)	2 (1.7)
MCN	43 (10.6)	33 (11.5)	10 (8.4)
NET/NEC	45 (11.1)	29 (10.1)	16 (13.4)
PDAC	67 (16.5)	37 (12.9)	30 (25.2)
SCN	110 (27.2)	86 (30.1)	24 (20.2)
SPN	60 (14.8)	42 (14.7)	18 (15.1)
Other	13 (3.2)	9 (3.1)	4 (3.4)
Margin				1
R0	400 (98.8)	282 (98.6)	118 (99.2)
R1	5 (1.2)	4 (1.4)	1 (0.8)
Delayed gastric emptying	23 (5.7)	12 (4.2)	11 (9.2)	0.046a

^aP < 0.05.

TO: Textbook outcome; IQR: Interquartile range; BMI: Body mass index; ASA: American Society of Anesthesiologists; EUS-FNA: Endoscopic ultrasound-guided fine-needle aspiration; CP/PPC: Chronic pancreatitis/pancreatic pseudocyst; IPMN: Intraductal papillary mucinous neoplasm; MCN: Mucinous cystic neoplasm; NET/NEC: Neuroendocrine tumor/neuroendocrine carcinoma; PDAC: Pancreatic ductal adenocarcinoma; SCN: Serous cystic neoplasm; SPN: Solid pseudopapillary neoplasm.

TO achievement and components

Overall, 286 of 405 patients (70.6%) achieved TO. Figure 1 illustrates the achievement rates for each TO component: The most commonly met criterion was the absence of in-hospital or 30-day mortality (98.5%), followed by no 30-day readmission (97.5%), absence of grade B/C PPH (96.3%), and no severe complications (87.2%). POPF of grade B/C was the least frequently avoided (80.7%), emerging as the primary challenge in achieving TO.

Figure 1 Textbook outcome after laparoscopic distal pancreatectomy, distributed by its individual components. PPH: Post-pancreatectomy hemorrhage; POPF: Post-operative pancreatic fistula.

Univariate and multivariate analyses of TO failure in LDP

Univariate analysis identified sex, cardiovascular disease, lesion consistency, splenic-vessel involvement, pancreatic portal hypertension, preoperative EUS-FNA, local invasion, and spleen preservation as factors associated with TO. Multivariate logistic regression retained four independent predictors (Table 2): Female sex remained as a protective factor [odds ratio (OR) = 0.62, 95% confidence interval (CI): 0.39-0.99, P = 0.046], whereas cystic-solid lesions (OR = 2.51, 95%CI: 1.34-4.69, P = 0.004), purely solid lesions (OR = 1.91, 95%CI: 1.06-3.44, P = 0.032), concomitant pancreatic portal hypertension (OR = 2.81, 95%CI: 1.06-7.45, P = 0.038) and preoperative EUS-FNA (OR = 2.66, 95%CI: 1.05-6.73, P = 0.038) were independently associated with an increased risk of TO failure.

Table 2 Univariate and multivariate analyses of failure to achieve textbook outcome in laparoscopic distal pancreatectomy.

Variables	Univariable OR (95%CI)	P value	Multivariable OR (95%CI)	P value
Age ≥ 50 years	1.44 (0.93-2.22)	0.100
Sex: Female	0.53 (0.34-0.82)	0.004a	0.62 (0.39-0.99)	0.046a
BMI grade
18.5-24	Ref.
< 18.5	0.84 (0.39-1.81)	0.658
≥ 24	1.50 (0.94-2.39)	0.088
Smoker or ex-smoker	1.35 (0.63-2.92)	0.439
Alcohol	1.55 (0.68-3.52)	0.297
Diabetes	0.58 (0.21-1.59)	0.293
Previous abdominal surgery	0.85 (0.39-1.89)	0.697
Asymptomatic	0.87 (0.56-1.34)	0.514
Anemia	1.24 (0.66-2.31)	0.508
Weight loss	1.68 (0.94-3.02)	0.082
ASA score
I and II	Ref.
III and IV	1.28 (0.72-2.28)	0.400
Cardiovascular disease	1.94 (1.09-3.45)	0.025	1.62 (0.88-2.98)	0.117
Pancreatitis	1.43 (0.66-3.11)	0.365
Lesion size (cm)	1.07 (0.99-1.16)	0.078
Lesion location
Tail	Ref.
Body	0.82 (0.52-1.29)	0.394
Neck	0.68 (0.32-1.45)	0.319
Lesion consistency
Cystic	Ref.
Cystic-solid	2.59 (1.43-4.69)	0.002a	2.51 (1.34-4.69)	0.004a
Solid	2.61 (1.52-4.47)	< 0.001a	1.91 (1.06-3.44)	0.032a
Splenic vessel involvement	1.84 (1.03-3.29)	0.041a	0.86 (0.41-1.81)	0. 689
Pancreatic portal hypertension	2.80 (1.27-6.16)	0.010a	2.81 (1.06-7.45)	0.038a
Pancreatic duct dilation	0.75 (0.42-1.34)	0.329
Preoperative EUS-FNA	2.81 (1.16-6.81)	0.022a	2.66 (1.05-6.73)	0.038a
Pancreatic texture
Hard	Ref.
Soft	1.11 (0.60-2.06)	0.746
Intraoperative blood transfusion	1.47 (0.52-4.13)	0.469
Local invasion	3.19 (1.23-8.29)	0.017a	1.97 (0.70-5.55)	0.200
Delayed gastric emptying	2.33 (1.00-5.43)	0.051
Spleen preservation	0.54 (0.34-0.84)	0.006a	0.83 (0.50-1.36)	0.455
Additional organ resection	2.43 (0.34-17.44)	0.378
Lymph node dissection	1.43 (0.93-2.20)	0.100
Malignancy	1.58 (0.96-2.61)	0.074

^aP < 0.05.

BMI: Body mass index; ASA: American Society of Anesthesiologists; EUS-FNA: Endoscopic ultrasound-guided fine-needle aspiration; OR: Odds ratio; CI: Confidence interval.

Development and validation of a risk-prediction model for TO failure

Utilizing the training cohort (n = 283), a risk-prediction model for TO failure was developed incorporating the four multivariate predictors sex, lesion consistency, pancreatic portal hypertension, and preoperative EUS-FNA and visualized as a nomogram (Figure 2). The C-indices were 0.661 (95%CI: 0.58-0.74) and 0.619 (95%CI: 0.53-0.71), respectively, indicating modest discrimination that marginally meets clinical acceptability. Calibration curves demonstrated close alignment between predicted and observed probabilities in both cohorts, with mean absolute errors of 0.020 and 0.022, respectively (Figure 3).

Figure 2 Nomogram for predicting textbook outcome failure after laparoscopic distal pancreatectomy. EUS-FNA: Endoscopic ultrasound-guided fine-needle aspiration; TO: Textbook outcome.

Figure 3 Calibration plots of the textbook outcome failure risk prediction model. A: Training set; B: Validation set.

DISCUSSION

LDP has established itself as the standard treatment modality for lesions affecting the pancreatic body and tail in high-volume medical centers. Despite this, the absence of a robust tool for comparing perioperative quality among institutions remains a notable gap in the field. A pressing need exists for a metric that enables the benchmarking of LDP quality across institutions. Since 2013, the composite measure termed “TO” has gained prominence in gastrointestinal, hepatobiliary, and pancreatic surgery as a comprehensive tool for assessing perioperative performance and guiding quality-improvement initiatives[15-18,22]. In pancreatic surgery specifically, TO has emerged as a critical instrument for addressing the limitations of single-indicator assessments. Nevertheless, current studies on TO display significant heterogeneity and imbalance.

The 2020 DPCG nationwide study was the first to introduce a TO specific to pancreatic surgery, requiring the absence of clinically significant POPF, bile leak, PPH (all grade B/C), severe complications (Clavien-Dindo ≥ III), 30-day readmission, and 30-day mortality[18]. This TO definition, validated in over 3000 open and laparoscopic PD and DP cases, includes a criterion (bile leak) relevant only to procedures involving biliary reconstruction, rendering it inapplicable to DP. In 2021, a United States multicenter study proposed a “best pancreatic surgery” framework similar to TO but excluded PPH, compromising its completeness[26]. More recently, a 2024 Spanish six-center study developed a DP-specific TO that excluded prolonged LOS, Clavien-Dindo grade ≥ III complications, in-hospital mortality, 90-day readmission, and POPF (grade B/C)[22]. However, this study combined open and laparoscopic cases and likewise ignored hemorrhage, limiting its ability to accurately depict the quality profile of LDP. Consequently, the existing body of evidence on TO predominantly concentrates on PD, with limited and inconsistent studies on DP. A dedicated TO investigation specifically focusing on a pure LDP cohort is currently lacking.

Building on the 2020 DPCG nationwide study and recognizing the absence of biliary anastomosis in LDP, we have refined the LDP-specific TO criteria to encompass the absence of POPF (grade B/C), PPH (grade B/C), severe morbidity (Clavien-Dindo ≥ III), 30-day readmission, and in-hospital or 30-day mortality[18]. Prolonged LOS was intentionally omitted due to its susceptibility to non-medical influences such as patient preference, bed availability, or insurance policies. By explicitly incorporating PPH, our definition addresses the predominant hemorrhagic complication specific to LDP, filling a crucial assessment gap and ensuring that TO accurately reflects operative technical quality and perioperative care effectiveness.

Among the 405 patients who underwent LDP in our study, 286 individuals (70.6%) achieved TO, surpassing the 67.4% reported for the LDP subgroup in the 2020 DPCG nationwide study[18] and the 58.2% recorded in a recent Spanish DP cohort[22]. This superior performance likely reflects both institutional volume and metric selection: Our institution performs over 80 LDP procedures annually, and the accrued expertise has been demonstrated to lower rates of intraoperative bleeding, POPF, and PPH. Moreover, by excluding prolonged LOS a variable influenced by external factors such as insurance policies and bed availability we circumvented the potential distortion of TO outcomes observed when LOS is included (the DPCG DP rate dropped from 67.4% to 63.3% after adding LOS), thereby providing a more accurate measure of surgical quality.

Univariate and multivariate analyses concordantly identified female sex, pre-operative EUS-FNA, pancreatic portal hypertension, and lesion consistency as independent predictors of TO failure following LDP.

Female sex emerged as a protective factor against TO failure in our cohort (OR = 0.62, 95%CI: 0.39-0.99, P = 0.046), consistent with the DPCG finding that female sex was a positive predictor of TO achievement (OR = 1.38, 95%CI: 1.01-1.90, P = 0.05)[18]. This protective association has been corroborated by larger datasets: A 2022 United States nationwide analysis of 1950 pancreatectomies showed that women more frequently attained an “ideal outcome” and outperformed men in all TO component (severe complications, 30-day mortality, re-operation, readmission)[27]. Similarly, the 2022 German pancreas registry (n = 10224) reported lower overall morbidity (57.3% vs 60.1%, P = 0.005), mortality (3.4% vs 4.9%, P < 0.001) and clinically relevant POPF after DP (22.5% vs 27.3%, P = 0.012) in females[28]. In our study, POPF rates were nearly halved in women (15.1% vs 25.6%, P = 0.009), while other major complications and 30-day mortality rates were comparable between genders. Given that POPF represents the primary barrier to TO after LDP, the favorable outcomes observed in females likely stem from an inherently lower risk of fistula formation. However, prospective multicenter investigations are warranted to establish causality.

At high-volume centers, EUS-FNA serves as an important preoperative diagnostic modality for both cystic and solid pancreatic lesions[29,30]. While EUS-FNA enhances diagnostic accuracy for pancreatic neoplasms with a low overall complication rate (3.4%)[30], it is an invasive maneuver that may induce local pancreatic inflammation or minor ductal injury, potentially increasing the risk of postoperative fistula formation. Additionally, complications such as hemorrhage, perforation, infection, and needle-tract seeding have been reported in specific cases[31,32]. In our series, patients subjected to preoperative EUS-FNA achieved TO in only 47.6% of cases, compared to a TO attainment rate of 71.9% in those without FNA (P = 0.017). Thus, the diagnostic advantages of EUS-FNA must be carefully weighed against its potential to impair TO after LDP.

Patients with concurrent pancreatic portal hypertension had a significantly higher risk of TO failure (OR = 2.94, 95%CI: 1.11-7.76, P = 0.030). Pancreatic portal hypertension commonly results from recurrent episodes of acute-on-chronic pancreatitis or direct tumor invasion of the splenic vein[33]. Chronic inflammation damages the endothelium of the splenic vein, leading to thrombosis and thickening of the vessel wall, while causing diffuse fibrosis of the pancreatic parenchyma with areas of edema or pseudocysts, significantly increasing tissue vulnerability[34]. This process also results in collateral dilation of the short gastric and left gastroepiploic veins, forming a delicate vascular network that is prone to bleeding during surgical dissection. Consequently, during LDP, the glandular transection plane becomes uneven and fragile, complicating secure closure with staples or sutures and heightening the risk of POPF; dilated collateral vessels are prone to bleeding with minimal manipulation, leading to diffuse hemorrhage or significant post-PPH; and the frequent necessity for concurrent vascular resection or extensive lymphadenectomy prolongs operative time and increases blood loss, diminishing the likelihood of achieving TO.

Patients with mixed cystic-solid (OR = 2.60, 95%CI: 1.40-4.82, P = 0.002) and solid lesions (OR = 1.99, 95%CI: 1.11-3.55, P = 0.020) were associated with markedly lower TO rates compared to those with purely cystic lesions. This disparity is likely attributable to the benign nature of cystic lesions (e.g., serous cystadenoma), which can often be excised with simple resection without lymph-node dissection, resulting in minimal surgical trauma and low complication rates[30]. In contrast, cystic-solid or solid lesions, such as solid pseudopapillary neoplasm, pancreatic adenocarcinoma, or neuroendocrine tumors, often necessitate extensive resection and lymph node clearance, leading to larger surgical fields and prolonged operative durations. Furthermore, the higher micro-vessel density in solid lesions heightens the risk of intraoperative bleeding, increasing the likelihood of POPF and intra-abdominal infections[35,36]. Therefore, tailored management based on the nature of the lesion is recommended: Cystic benign tumors are optimally addressed through laparoscopic spleen-preserving LDP to minimize trauma, whereas cystic-solid or solid tumors require accurate preoperative staging, meticulous vascular control, and secure pancreatic-stump closure to reduce morbidity and enhance TO achievement rates.

Additionally, utilizing the independent predictors of TO failure identified through multivariable analysis sex, preoperative EUS-FNA, pancreatic portal hypertension, and lesion consistency we developed and internally validated a visual nomogram risk model. As a simple and objective preoperative assessment tool, the nomogram provides an individualized estimate of the risk of TO failure, guiding tailored interventions and optimizing perioperative pathways. The nomogram achieved modest discrimination in both the training set (C-index = 0.661) and validation set (C-index = 0.619), only marginally meeting the threshold of clinical acceptability. Although calibration plots indicated good agreement (mean absolute error < 0.03), the limited C-index highlights the need for additional predictors, advanced modeling techniques (e.g., machine learning), and external validation before the model can be recommended for routine clinical use.

Importantly, it should be acknowledged that TO was conceived as a surrogate indicator for perioperative technical quality rather than a reflection of patient-centered endpoints such as quality of life, pain, fatigue, or time to functional recovery. Surgeons traditionally track mortality and morbidity, whereas administrators emphasize cost and LOS; patients, however, prioritize symptom relief, return to normal activities, and long-term well-being. Consequently, over-interpreting TO as a comprehensive measure of “surgical success” risks creating an illusion of excellence while neglecting outcomes that matters most to patients. Future prospective studies should integrate patient-reported outcome measures and functional recovery scores to better align quality assessment with patient priorities and to clarify whether achieving TO truly translates into improved quality of life after LDP.

This study has several limitations. Firstly, being a single-center retrospective cohort, it is susceptible to selection bias, necessitating caution when generalizing the findings to community or low-volume centers. Subsequent research should encompass external cohorts from diverse regions and institutions with varying case volumes to validate the applicability of our TO criteria and risk factors. Secondly, the model development and validation were performed within the same institution, lacking a fully independent external dataset and potentially overestimating predictive performance. Prospective, multicenter registries are essential for further calibration and refinement. Lastly, the evaluation of TO only encompassed 30-day postoperative outcomes, excluding long-term survival, tumor recurrence, and quality-of-life assessments. Consequently, the long-term effectiveness and patient outcomes following LDP remain incompletely characterized.

CONCLUSION

In this single-institution cohort of 405 patients, we established a tailored TO for LDP, defined as the absence of grade B/C POPF, grade B/C PPH, Clavien-Dindo ≥ III morbidity, 30-day readmission, in-hospital, and 30-day mortality. TO was achieved in 70.6% of cases, serving as a robust standard for perioperative quality assessment. Our analysis identified female sex as a protective factor, while preoperative EUS-FNA, pancreatic portal hypertension, and cystic-solid or solid lesion consistency independently predicted TO failure. Leveraging these insights, we developed and internally validated a nomogram that translates these factors into individualized risk estimations, providing a practical and standardized tool for preoperative risk assessment and ongoing quality enhancement in LDP.

ACKNOWLEDGEMENTS

We acknowledge all patients who contributed to this study.