A biomarker-integrated nomogram for predicting postoperative pancreatic fistula in low- to intermediate-risk patients after pancreaticoduodenectomy

Introduction

Postoperative pancreatic fistula (POPF) is a frequent complication after pancreaticoduodenectomy (PD) and associated with poor outcomes, or even death (1,2). Despite advances in understanding the multifactorial pathogenesis of POPF, encompassing anatomical, surgical, and biological determinants, no universally effective strategy exists to prevent this complication (3-6). In this context, early and accurate prediction of POPF assumes critical importance, enabling timely escalation of monitoring and personalized adjustment of prophylactic measures.

The Fistula Risk Score (FRS) is the most widely recognized and applied predictive score, which features intraoperative assessment and assigns patients into four risk strata (negligible risk, low risk, intermediate risk and high risk) (7). The implementation of FRS and its derivative versions in clinical practice has significantly modified management protocols (8-10). Nevertheless, the FRS framework relies exclusively on static anatomical and intraoperative variables and fails to account for the dynamic biological processes governing POPF evolution.

Serum biomarkers, such as procalcitonin (PCT) and C-reactive protein (CRP), have emerged as vital predictors for postoperative complications across surgical disciplines (11-16). PCT, a peptide precursor elevated in bacterial infections, peaks within 48–72 hours postoperatively and correlates with septic complications (17), while CRP serves as a sensitive marker of tissue inflammation and injury (18). Despite their prognostic potential, these biomarkers are seldomly included in existing POPF prediction models, representing a critical gap in dynamic risk assessment.

The nomogram has emerged in recent years as an effective and practical tool for predicting survival outcomes and postoperative complications (19-21). By transforming complex statistical models into intuitive, point-based visual formats, the nomogram allows clinicians to estimate patient-specific probabilities of adverse events, thereby serving as valuable aids in clinical decision-making (22).

Thus, the aim of this study was to evaluate the diagnostic accuracy of serum biomarkers and then develop a risk prediction model for POPF in patients with low to intermediate risk after PD. We present this article in accordance with the TRIPOD reporting checklist (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-1-0008/rc) (23).

Methods

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Ethics Committee on Clinical Investigation of Nanjing Drum Tower Hospital (approval No. 2019-115-01) and individual consent for this retrospective analysis was waived.

Study design and population

From January 2019 to May 2021, we collected clinical data from patients who underwent PD in the Department of Pancreatic and Metabolic Surgery at Nanjing Drum Tower Hospital. Patients were stratified according to the FRS. Patients classified as low- and intermediate-risk categories were included in the analysis. Exclusion criteria included: (I) incomplete clinical documentation; (II) a previous ongoing infection; (III) emergency surgical intervention; and (IV) comorbid conditions potentially confounding PCT interpretation, including but not limited to medullary thyroid carcinoma and severe renal insufficiency. The sample size was determined by the number of consecutive eligible patients who underwent PD during the study period.

Data collection, outcomes and definition

Patient demographic characteristics and preoperative laboratory test findings were collected, including age, sex, body mass index (BMI), presence of diabetes, total bilirubin, albumin, prealbumin and hemoglobin. Nutrition status was assessed by special clinical pharmacist using Nutrition Risk Screening (NRS) 2002 (24). Operations were performed by four teams under the same ward. Conventional PD and pylorus-preserving PD were both allowed according to the surgeon’s preferences, although the former is in the majority. Reconstruction between the pancreatic stump and the gastrointestinal tract was performed using pancreaticojejunostomy and duct-to-mucosa technique with a pancreatic duct stent. Two passive drainage tubes were routinely placed in the vicinity of the pancreatic and biliary anastomosis. High-risk pathology was defined as ampullary, duodenal, cystic, or islet cell pathology, whereas pancreatic adenocarcinoma or chronic pancreatitis, or those with firm pancreatic parenchyma were classified as low-risk pathology, in accordance with the FRS.

Intraoperative evaluation of the four parameters enabled risk stratification into clinically defined categories: negligible risk (0 points), low risk (1–2 points), intermediate risk (3–6 points), and high risk (7–10 points) according to FRS.

White blood cell count (WBC), CRP and PCT tests were carried out on postoperative day (POD) 1, 3, 5. The concentration of amylase in drain fluid was measured routinely on POD1, 3, 5 and the highest one was recorded in case of multiple drains. Prophylactic somatostatin or somatostatin analogs and antibiotics were administered during postoperative period.

POPF was defined in line with the 2016 International Study Group of Pancreatic Fistula (ISGPS) definition (25), and it should be noted that a biochemical leak, the former grade A POPF, was not included as a true pancreatic fistula. Grade B POPF included cases requiring prolonged drainage, antibiotics, or interventional procedures, while grade C POPF required reoperation or resulted in organ failure. Definitions of other postoperative complications also refer to current evidence (26-28).

Statistical analysis

Statistical analysis was performed using SPSS 23.0 (SPSS Inc., Chicago, IL, USA) and R. Quantitative outcomes were presented as median and interquartile range (IQR) or mean and standard deviation, and compared with Student’s t-tests. Qualitative outcomes were presented as percentages and compared with χ² tests. Receiver operating characteristic (ROC) curves were performed to assess diagnostic accuracy of PCT and CRP, and the optimal cut-off value was determined by Youden’s index for further logistic regression analysis.

Variables with a P value <0.25 in univariate analysis and clinically relevant variables, despite their P value exceeding 0.25, were retained for multivariable logistic regression. In the final multivariate model, variables achieving statistical significance (P<0.05) were identified as independent risk factors and subsequently incorporated into a nomogram using R. The nomogram was designed to visualize the contribution of each variable by assigning weighted points proportional to their regression coefficients. Model performance of discrimination and calibration was assessed using areas under the ROC curve (AUC) and calibration curve. Internal validation was performed using bootstrap resampling with 1,000 iterations to assess the model’s stability and reduce overfitting. Clinical utility of the nomogram was assessed using decision curve analysis (DCA). Patients with missing data were excluded from the analysis, and a complete-case analysis was performed.

Results

Patient characteristics and complication

A total of 246 consecutive patients undergoing PD between January 2019 and May 2021 were initially screened. After excluding negligible-risk patients (n=3), high-risk patients (n=37), patients with missing data (n=5), patients with an ongoing infection (n=1), patients undergoing emergency operation (n=5), 195 patients with low or intermediate risk of POPF, 51 (26.2%) and 144 (73.8%) respectively, were included in the final cohort.

Patients’ baseline information was summarized in Table 1. The median age was 63.0 years and male made up more than half of the study population (117 male, 60.0% vs. 78 female, 40.0%). Diabetes was present in 20.0% of the patients. Unintentional weight loss >10% within 6 months before surgery was observed in 37 (19.0%) patients. Fifty-eight (29.7%) patients received preoperative biliary drainage.

Among patients undergoing PD with low or intermediate risk, 60 (30.8%) developed POPF. Grade B and C POPF occurred in 55 (28.2%) and 5 (2.6%) patients. Of the 5 patients who developed grade C POPF, they all required reoperation, 1 for POPF associated hemorrhage, 4 for uncontrollable POPF and intra-abdominal infections.

PCT, CRP and WBC levels and diagnosis of POPF

The level of PCT, CRP and WBC in patients with and without POPF are shown in Figure 1. The median PCT concentration on POD1, 3, 5 was respectively 0.223 (IQR, 0.121–0.481), 0.224 (IQR, 0.117–0.468), 0.130 (IQR, 0.080–0.250) ng/mL in patients without POPF vs. 0.380 (IQR, 0.187–0.892), 0.546 (IQR, 0.342–1.243), 0.290 (IQR, 0.196–0.570) ng/mL in patients with POPF (P=0.31 on POD1, P=0.02 on POD3 and P=0.07 on POD5 for difference). The median CRP concentration on POD1, 3, 5 was respectively 58.8 (IQR, 44.9–73.5), 81.8 (IQR, 57.8–108.4), 52.0 (IQR, 25.1–78.4) mg/L in patients without POPF vs. 69.2 (IQR, 49.8–86.3), 131.3 (IQR, 90.9–187.1), 93.5 (IQR, 70.5–121.3) mg/L in patients with POPF (P=0.08 on POD1, P<0.001 on POD3 and P<0.001 on POD5 for difference). The median WBC concentration on POD1, 3, 5 was respectively 11.9 (IQR, 9.6–15.1)×10⁹/L, 10.2 (IQR, 7.8–13.1)×10⁹/L, 9.4 (IQR, 7.2–11.3)×10⁹/L in patients without POPF vs. 11.9 (IQR, 8.5–14.7)×10⁹/L, 10.5 (IQR, 6.8–12.6)×10⁹/L, 9.3 (IQR, 7.0–10.3)×10⁹/L in patients with POPF (P=0.42 on POD1, P=0.60 on POD3 and P=0.72 on POD5 for difference).

Figure 1 Comparison of PCT (A), CRP (B) and WBC (C) levels between patients with and without POPF. Levels of PCT and CRP on POD3 and POD5 were significantly higher in patients who developed POPF. Although the difference was not statistically significant, PCT levels on POD5 tended to be higher in patients with POPF compared to those without (P=0.07). Data are presented as median and interquartile range. ns, not significant; *, P<0.05; ***, P<0.001. CRP, C-reactive protein; PCT, procalcitonin; POD, postoperative day; POPF, postoperative pancreatic fistula; WBC, white blood cell count.

The kinetics of PCT and CRP was similar, with an early peak and a subsequently decrease, in both POPF and no POPF group, reaching their peak on POD3.

PCT and CRP demonstrated good diagnostic accuracy on both POD3 and POD5 (Figure 2). On POD3 the AUC for PCT and CRP was 0.7573 [95% confidence interval (CI): 0.6886–0.8259] and 0.7594 (95% CI: 0.6916–0.8272). On POD5 the AUC for PCT and CRP was 0.7857 (95% CI: 0.7170–0.8544) and 0.7928 (95% CI: 0.7240–0.8617). Considering the importance of early diagnosis, POD3 values of PCT and CRP were selected for subsequent analyses. On the basis of the ROC analysis, the optimal diagnostic cut-off value for PCT and CRP on POD3 was 0.229 ng/mL and 111.15 mg/L respectively. Under the condition of optimal cut-off value, the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) for POPF was 95.0%, 51.9%, 46.7%, 95.9% for PCT, and 65.0%, 79.3%, 58.2%, 83.6% for CRP. Logistic regression analysis confirmed POD3 PCT and CRP as powerful predictors of POPF after PD in both univariate and multivariate analyses (Table 2).

Figure 2 ROC curves of PCT (A) and CRP (B) on POD1, 3, 5. On POD3, the AUC is 0.7573 (95% CI: 0.6886–0.8259) for PCT and 0.7594 (95% CI: 0.6916–0.8272) for CRP; on POD5, the AUC is 0.7857 (95% CI: 0.7170–0.8544) and 0.7928 (95% CI: 0.7240–0.8617), respectively. AUC, area under the curve; CI, confidence interval; CRP, C-reactive protein; PCT, procalcitonin; POD, postoperative day; ROC, receiver operating characteristic.

Model development

In the univariate logistic regression analysis, variables meeting the prespecified threshold of P<0.25 (age, male gender, BMI, weight loss, high-risk pathology, soft pancreatic texture, Wirsung diameter, POD3 PCT >0.229 ng/mL and POD3 CRP >111.15 mg/L) along with clinically relevant variable (intraoperative blood loss) were included in the multivariate analysis. Multivariate analysis identified high-risk pathology, soft pancreatic texture, Wirsung diameter, POD3 PCT >0.229 ng/mL and POD3 CRP >111.15 mg/L as independent risk factors for POPF, and we subsequently developed a predictive nomogram for POPF (Figure 3). To facilitate clinical implementation, a points-to-risk conversion table translating the total nomogram score into the predicted probability of POPF is provided in Table S1.

Figure 3 Nomogram for predicting POPF following PD in patients at low to intermediate risk. The model incorporates five predictors: high-risk pathology, soft pancreatic texture, Wirsung diameter, POD3 PCT >0.229 ng/mL and POD3 CRP >111.15 mg/L. To estimate an individual’s risk of POPF, points are assigned for each variable based on its location on the corresponding scale; the sum is then mapped to the total points axis to determine the predicted probability of fistula. CRP, C-reactive protein; PCT, procalcitonin; PD, pancreaticoduodenectomy; POD, postoperative day; POPF, postoperative pancreatic fistula.

Model performance and clinical utility

The model demonstrated excellent discrimination with an AUC of 0.9374 (95% CI: 0.9056–0.9692) (Figure 4). Bootstrap-corrected calibration curves showed close alignment between predicted and observed POPF probabilities (Figure 5). DCA showed that the nomogram provided higher net benefit across a wide range of threshold probabilities compared to the “treat all” or “treat none” strategies (Figure 6).

Figure 4 ROC curve for the predictive nomogram. The model demonstrated excellent discrimination with an AUC of 0.9374 (95% CI: 0.9056–0.9692). AUC, areas under the curve; CI, confidence interval; ROC, receiver operating characteristic.

Figure 5 Calibration curve for the predictive nomogram. The x-axis represents the predicted probability of POPF, and the y-axis represents the actual observed incidence. The 45-degree diagonal line indicates perfect calibration. The apparent curve represents the original model performance, while the bias-corrected curve reflects the performance after 1,000 bootstrap resampling iterations. POPF, postoperative pancreatic fistula.

Figure 6 DCA for the predictive nomogram. The net benefit is plotted against threshold probabilities. The nomogram demonstrates a higher net benefit than either the treat-all or treat-none strategies across a wide range of thresholds. DCA, decision curve analysis.

Discussion

In our study, we evaluated the clinical utility of postoperative serum biomarkers in patients undergoing PD and developed a nomogram that integrates PCT, CRP, and selected components of the FRS to predict the risk of POPF in patients classified as low to intermediate risk.

Despite advances in perioperative care, POPF remains a devastating complication after PD, with reported incidence rates ranging from 10% to 30%, or even more (1,7). Advancements in predicting POPF have played a crucial role in refining surgical strategies and decision-making (29-32). The FRS has earned widespread acceptance by stratifying patients on the basis of intraoperative findings and informing clinical pathways; however, its exclusive reliance on static, intraoperative variables may overlook the dynamic biological processes driving POPF development. Giving the prediction of POPF should be a dynamic process spanning the entire perioperative period, our nomogram uniquely incorporates both intraoperative parameters (highrisk pathology, soft gland texture, Wirsung duct diameter) and postoperative serum biomarkers (POD3 PCT >0.229 ng/mL and CRP >111.15 mg/L), thus capturing the evolving inflammatory milieu that underpins anastomotic failure. Previous studies have validated the predictive role of serum biomarkers on the early diagnosis of postoperative infectious complications (13,14,33). Giardino et al. found in their prospective study that PCT and CRP were highly predictive of complications on POD1–3 (13). Partelli et al. reported that CRP on POD2 >150 mg/L [odds ratio (OR) =3.480, P=0.021] and CRP on POD3 >185 mg/L (OR =6.738, P=0.003) were independent predictors of clinically relevant POPF (33). In our study, both POD3 and POD5 serum biomarkers exhibited favorable diagnostic efficacy in predicting POPF. We opted to focus on POD3 measurements because evaluating biomarkers on POD3 allows for earlier identification of patients at risk for POPF, providing clinicians with a critical window to implement timely and individualized postoperative interventions. Moreover, numerous high-volume pancreatic surgery centers consider POD3 as the standard time point for early drain removal (34,35), promoting consistency across studies and clinical practices. Serum biomarkers on POD1 demonstrated limited diagnostic efficacy, and this limitation may stem from early postoperative physiological responses, including systemic inflammation induced by surgical trauma and potential intraoperative bacterial translocation, which can lead to nonspecific elevations in serum biomarker levels. Highrisk pathology, soft gland texture and small Wirsung duct diameter have been consistently identified as well-established predictors of POPF across multicenter cohorts and validated risk stratification systems (7,10). Notably, intraoperative blood loss did not emerge as an independent predictor in our cohort—consistent with some reports (10,36,37)—and may reflect the inherent difficulty in accurately quantifying blood loss or its limited mechanistic role in fistula formation.

Current clinical guidelines emphasize the significance of tailoring postoperative management approaches to individual patient characteristics after PD, aiming to optimize outcomes and reduce complications. Previous studies support the safety of early drain removal after PD in carefully selected patient populations (34,35,38). Bassi et al. reported that early drain removal (≥3 d) significantly decreased the rates of POPF (P=0.0001), abdominal (P=0.002), and pulmonary (P=0.007) complications compared to standard drain removal (≥5 d), in patients deemed at low risk for POPF, as indicated by a drain amylase level ≤5,000 U/L on POD1 (34). Dai et al. in their multicenter randomized controlled trial (RCT) concluded that early drain removal is safe in patients with low or intermediate risk of POPF according to the FRS (35). However, it is important to note that high-risk patients were not excluded in the analysis, comprising approximately 3.2% of the study population. Previous meta-analyses evaluating the efficacy of somatostatin and its analogues in reducing postoperative complications following pancreatic surgery have reported conflicting outcomes (39,40). Gurusamy et al. in their study identified 19 RCT trials involving 2,245 patients, and their results showed that somatostatin analogues reduced the incidence of POPF [relative risk (RR) =0.63, 95% CI: 0.52–0.77] and postoperative complications (RR =0.69, 95% CI: 0.60–0.79) (39). On the contrary, the other meta-analysis encompassing 12 RCT focusing on PD revealed that prophylactic administration of somatostatin and its analogues did not significantly reduce the incidence of POPF (OR =0.73, 95% CI: 0.51–1.05, P=0.09) or postoperative mortality (OR =1.78, 95% CI: 0.94–3.39, P=0.08) (40). Given the difficulty in treating POPF and its associated high mortality, surgical drains and the administration of somatostatin analogues are still commonly utilized in postoperative protocols to mitigate these risks. Our study concentrates on patients classified as low to intermediate risk according to the FRS. By restricting the cohort to patients with low to intermediate FRS, spectrum bias may have been introduced, which could limit generalizability. However, this population represents a clinically relevant group in whom postoperative management strategies remain controversial. Our nomogram exhibited excellent discrimination, calibration, and clinical utility. The DCA indicated the clinical usefulness of the nomogram, suggesting that effective preventive measures provided greater clinical benefits than all or no treatment when the threshold probability ranges from 0% to 80%. For example, a patient with low to intermediate FRS who presents with a non-high-risk pathology, a pancreatic duct diameter ≥3 mm, firm gland texture, and POD3 PCT and CRP levels below the identified thresholds would yield a low predicted probability of POPF on the nomogram, potentially supporting safe early drain removal on POD5, or even POD3 in carefully selected cases.

There are several limitations in our study. First, recent studies have highlighted the predictive value of other postoperative biochemical parameters such as drain fluid amylase, serum amylase, and serum lipase in the early identification of POPF (29-31,41). Future studies may consider integrating a broader panel of biochemical parameters to improve diagnostic performance. Second, our study was a retrospective analysis conducted at a single high-volume center, which may limit the generalizability of the findings due to potential institutional biases in patient selection, surgical technique, and perioperative management. Moreover, the absence of external validation restricts the model’s applicability to other clinical settings. Further prospective, multicenter studies with external validation cohorts are needed to confirm the robustness and clinical utility of the nomogram across broader populations.

Conclusions

In this study, we demonstrated that PCT and CRP have good diagnostic accuracy for POPF after PD. Based on these findings, we developed a nomogram that integrates intraoperative factors with early postoperative serum biomarkers for patients with low to intermediate risk of POPF. This dynamic and individualized risk assessment model may assist clinicians in guiding postoperative decision-making and improve clinical outcomes.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-1-0008/rc

Data Sharing Statement: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-1-0008/dss

Peer Review File: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-1-0008/prf

Funding: This work was supported by the Clinical Medical Expert Team Introduction Project of Xuzhou, China (grant No. 2018TD005) and the Xuzhou Science and Technology Council (grant No. KC23174).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-1-0008/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Nanjing Drum Tower Hospital (approval No. 2019-115-01), and individual consent for this retrospective analysis was waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Allen PJ, Gönen M, Brennan MF, et al. Pasireotide for postoperative pancreatic fistula. N Engl J Med 2014;370:2014-22. [Crossref] [PubMed]
2. Fuks D, Piessen G, Huet E, et al. Life-threatening postoperative pancreatic fistula (grade C) after pancreaticoduodenectomy: incidence, prognosis, and risk factors. Am J Surg 2009;197:702-9. [Crossref] [PubMed]
3. Kawaida H, Kono H, Hosomura N, et al. Surgical techniques and postoperative management to prevent postoperative pancreatic fistula after pancreatic surgery. World J Gastroenterol 2019;25:3722-37. [Crossref] [PubMed]
4. Wu J, Zhang G, Yao X, et al. Achilles'heel of laparoscopic pancreatectomy: reconstruction of the remnant pancreas. Expert Rev Gastroenterol Hepatol 2020;14:527-37. [Crossref] [PubMed]
5. Marchegiani G, Barreto SG, Bannone E, et al. Postpancreatectomy Acute Pancreatitis (PPAP): Definition and Grading From the International Study Group for Pancreatic Surgery (ISGPS). Ann Surg 2022;275:663-72. [Crossref] [PubMed]
6. Chen H, Wang W, Fu N, et al. Characterization of Pancreatic Fistula After Post-pancreatectomy Acute Pancreatitis. Ann Surg 2025;282:1045-51. [Crossref] [PubMed]
7. Callery MP, Pratt WB, Kent TS, et al. A prospectively validated clinical risk score accurately predicts pancreatic fistula after pancreatoduodenectomy. J Am Coll Surg 2013;216:1-14. [Crossref] [PubMed]
8. Miller BC, Christein JD, Behrman SW, et al. A multi-institutional external validation of the fistula risk score for pancreatoduodenectomy. J Gastrointest Surg 2014;18:172-79; discussion 179-80. [Crossref] [PubMed]
9. Tang B, Lin Z, Ma Y, et al. A modified alternative fistula risk score (a-FRS) obtained from the computed tomography enhancement pattern of the pancreatic parenchyma predicts pancreatic fistula after pancreatoduodenectomy. HPB (Oxford) 2021;23:1759-66. [Crossref] [PubMed]
10. Mungroop TH, van Rijssen LB, van Klaveren D, et al. Alternative Fistula Risk Score for Pancreatoduodenectomy (a-FRS): Design and International External Validation. Ann Surg 2019;269:937-43. [Crossref] [PubMed]
11. Giaccaglia V, Salvi PF, Antonelli MS, et al. Procalcitonin Reveals Early Dehiscence in Colorectal Surgery: The PREDICS Study. Ann Surg 2016;263:967-72. [Crossref] [PubMed]
12. Limper M, de Kruif MD, Duits AJ, et al. The diagnostic role of procalcitonin and other biomarkers in discriminating infectious from non-infectious fever. J Infect 2010;60:409-16. [Crossref] [PubMed]
13. Giardino A, Spolverato G, Regi P, et al. C-Reactive Protein and Procalcitonin as Predictors of Postoperative Inflammatory Complications After Pancreatic Surgery. J Gastrointest Surg 2016;20:1482-92. [Crossref] [PubMed]
14. Mintziras I, Maurer E, Kanngiesser V, et al. C-reactive protein and drain amylase accurately predict clinically relevant pancreatic fistula after partial pancreaticoduodenectomy. Int J Surg 2020;76:53-8. [Crossref] [PubMed]
15. Chen H, Wang C, Xia W, et al. Early prediction of post-pancreatectomy acute pancreatitis after pancreaticoduodenectomy based on serum C-reactive protein. Pancreatology 2025;25:208-13. [Crossref] [PubMed]
16. Chuang CL, Yeh HT, Niu KY, et al. Diagnostic performances of procalcitonin and C-reactive protein for sepsis: a systematic review and meta-analysis. Eur J Emerg Med 2025;32:248-58. [Crossref] [PubMed]
17. Meisner M, Tschaikowsky K, Hutzler A, et al. Postoperative plasma concentrations of procalcitonin after different types of surgery. Intensive Care Med 1998;24:680-4. [Crossref] [PubMed]
18. Straatman J, Harmsen AM, Cuesta MA, et al. Predictive Value of C-Reactive Protein for Major Complications after Major Abdominal Surgery: A Systematic Review and Pooled-Analysis. PLoS One 2015;10:e0132995. [Crossref] [PubMed]
19. Balachandran VP, Gonen M, Smith JJ, et al. Nomograms in oncology: more than meets the eye. Lancet Oncol 2015;16:e173-80. [Crossref] [PubMed]
20. Cai Z, Lin H, Li Z, et al. A prediction nomogram for postoperative gastroparesis syndrome in right colon cancer: a retrospective study. Langenbecks Arch Surg 2023;408:148. [Crossref] [PubMed]
21. Lim JH, Han A, Cho SJ, et al. Nomogram Prediction for Gastric Cancer Development. Clin Transl Gastroenterol 2025;16:e00833. [Crossref] [PubMed]
22. Tong C, Miao Q, Zheng J, et al. A novel nomogram for predicting the decision to delayed extubation after thoracoscopic lung cancer surgery. Ann Med 2023;55:800-7. [Crossref] [PubMed]
23. Collins GS, Reitsma JB, Altman DG, et al. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD statement. BMJ 2015;350:g7594. [Crossref] [PubMed]
24. Kondrup J, Allison SP, Elia M, et al. ESPEN guidelines for nutrition screening 2002. Clin Nutr 2003;22:415-21. [Crossref] [PubMed]
25. Bassi C, Marchegiani G, Dervenis C, et al. The 2016 update of the International Study Group (ISGPS) definition and grading of postoperative pancreatic fistula: 11 Years After. Surgery 2017;161:584-91. [Crossref] [PubMed]
26. Wente MN, Veit JA, Bassi C, et al. Postpancreatectomy hemorrhage (PPH): an International Study Group of Pancreatic Surgery (ISGPS) definition. Surgery 2007;142:20-5. [Crossref] [PubMed]
27. 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;142:761-8. [Crossref] [PubMed]
28. Koch M, Garden OJ, Padbury R, et al. Bile leakage after hepatobiliary and pancreatic surgery: a definition and grading of severity by the International Study Group of Liver Surgery. Surgery 2011;149:680-8. [Crossref] [PubMed]
29. Bannone E, Marchegiani G, Vollmer C, et al. Postoperative Serum Hyperamylasemia Adds Sequential Value to the Fistula Risk Score in Predicting Pancreatic Fistula after Pancreatoduodenectomy. Ann Surg 2023;278:e293-301. [Crossref] [PubMed]
30. Rykina-Tameeva N, MacCulloch D, Hipperson L, et al. Drain fluid biomarkers for the diagnosis of clinically relevant postoperative pancreatic fistula: a diagnostic accuracy systematic review and meta-analysis. Int J Surg 2023;109:2486-99. [Crossref] [PubMed]
31. Caputo D, Angeletti S, Ciccozzi M, et al. Role of drain amylase levels assay and routinary postoperative day 3 abdominal CT scan in prevention of complications and management of surgical drains after pancreaticoduodenectomy. Updates Surg 2020;72:727-41. [Crossref] [PubMed]
32. Kaiser JD, Bräuherr F, Biesel EA, et al. Preoperative prediction of postoperative pancreatic fistula after Pancreaticoduodenectomy: Determination and validation of a cut-off value for the Roberts Score. Am J Surg 2025;245:116356. [Crossref] [PubMed]
33. Partelli S, Pecorelli N, Muffatti F, et al. Early Postoperative Prediction of Clinically Relevant Pancreatic Fistula after Pancreaticoduodenectomy: usefulness of C-reactive Protein. HPB (Oxford) 2017;19:580-6. [Crossref] [PubMed]
34. Bassi C, Molinari E, Malleo G, et al. Early versus late drain removal after standard pancreatic resections: results of a prospective randomized trial. Ann Surg 2010;252:207-14. [Crossref] [PubMed]
35. Dai M, Liu Q, Xing C, et al. Early Drain Removal is Safe in Patients With Low or Intermediate Risk of Pancreatic Fistula After Pancreaticoduodenectomy: A Multicenter, Randomized Controlled Trial. Ann Surg 2022;275:e307-14. [Crossref] [PubMed]
36. Shubert CR, Wagie AE, Farnell MB, et al. Clinical Risk Score to Predict Pancreatic Fistula after Pancreatoduodenectomy: Independent External Validation for Open and Laparoscopic Approaches. J Am Coll Surg 2015;221:689-98. [Crossref] [PubMed]
37. Grendar J, Jutric Z, Leal JN, et al. Validation of Fistula Risk Score calculator in diverse North American HPB practices. HPB (Oxford) 2017;19:508-14. [Crossref] [PubMed]
38. Kawai M, Tani M, Terasawa H, et al. Early removal of prophylactic drains reduces the risk of intra-abdominal infections in patients with pancreatic head resection: prospective study for 104 consecutive patients. Ann Surg 2006;244:1-7. [Crossref] [PubMed]
39. Gurusamy KS, Koti R, Fusai G, et al. Somatostatin analogues for pancreatic surgery. Cochrane Database Syst Rev 2012;CD008370.
40. Adiamah A, Arif Z, Berti F, et al. The Use of Prophylactic Somatostatin Therapy Following Pancreaticoduodenectomy: A Meta-analysis of Randomised Controlled Trials. World J Surg 2019;43:1788-801. [Crossref] [PubMed]
41. Tang BJ, Li SJ, Wang PF, et al. Predictive value of postoperative serum lipase level for postoperative pancreatic fistula after pancreaticoduodenectomy. Hepatobiliary Pancreat Dis Int 2025;24:197-205. [Crossref] [PubMed]