A simple nomogram for early postoperative risk prediction of clinically relevant pancreatic fistula after pancreatoduodenectomy

Optimizing prediction models for pancreatic fistula after pancreatectomy: Current status and future perspectives

Postoperative pancreatic fistula (POPF) is a frequent complication after pancreatectomy, leading to increased morbidity and mortality. Optimizing prediction models for POPF has emerged as a critical focus in surgical research. Although over sixty models following pancreaticoduodenectomy, predominantly reliant on a variety of clinical, surgical, and radiological parameters, have been documented, their predictive accuracy remains suboptimal in external validation and across diverse populations. As models after distal pancreatectomy continue to be progressively reported, their external validation is eagerly anticipated. Conversely, POPF prediction after central pancreatectomy is in its nascent stage, warranting urgent need for further development and validation. The potential of machine learning and big data analytics offers promising prospects for enhancing the accuracy of prediction models by incorporating an extensive array of variables and optimizing algorithm performance. Moreover, there is potential for the development of personalized prediction models based on patient- or pancreas-specific factors and postoperative serum or drain fluid biomarkers to improve accuracy in identifying individuals at risk of POPF. In the future, prospective multicenter studies and the integration of novel imaging technologies, such as artificial intelligence-based radiomics, may further refine predictive models. Addressing these issues is anticipated to revolutionize risk stratification, clinical decision-making, and postoperative management in patients undergoing pancreatectomy.

Machine learning approach to predict postpancreatectomy hemorrhage following pancreaticoduodenectomy: a retrospective study

BACKGROUND

Postpancreatectomy hemorrhage (PPH) is a rare yet dreaded complication following pancreaticoduodenectomy (PD). This retrospective study aimed to explore a machine learning (ML) model for predicting PPH in PD patients.

METHODS

A total of 284 patients who underwent open PD at our institute were included in the analysis. To address the issue of imbalanced data, the adaptive synthetic sampling (ADASYN) technique was employed. The best-performing ML model was selected using the PyCaret library in Python and evaluated based on recall, precision, and F1 score metrics. In addition to assessing the model's performance on the test data, bootstrap validation (n = 1000) with the original dataset was conducted.

RESULTS

PPH occurred in 11 patients (3.9%), with a median onset time of 22 days postoperatively. These minority cases were oversampled to 85 using ADASYN. The extra trees classifier demonstrated superior performance with recall, precision, and F1 score of 0.967, 0.914, and 0.937, respectively. Both validation using the test data and bootstrap resampling consistently demonstrated recall, precision, and F1 score exceeding 0.9. The model identified the peak value of C-reactive protein during the first 7 postoperative days as the most significant feature, followed by the preoperative neutrophil-to-lymphocyte ratio.

CONCLUSIONS

This study highlights the potential of the ML approach to predict PPH occurrence following PD. Vigilance and early interventions guided by such model predictions could positively impact outcomes for high-risk patients.

Serum Creatinine and Amylase in Drain to Predict Pancreatic Fistula Risk after Pancreatoduodenectomy

INTRODUCTION

The identification of patients with low risk of clinically relevant postoperative pancreatic fistula (CR-POPF) and postoperative hemorrhage (PPH) can guide drain removal after pancreatoduodenectomy (PD). However, drain fluid amylase (DFA) ≤5,000 U/L on postoperative day (POD) 1 does not robustly predict the absence of CR-POPF.

METHODS

Consecutive patients undergoing PD at Sun Yat-sen University Cancer Center between July 2018 and October 2021 were analyzed. Recursive partitioning analysis was used to classify patients into groups with different risks of CR-POPF and PPH.

RESULTS

Among 288 consecutive patients included, 99 patients (34.38%) developed CR-POPF (86 grade B and 13 grade C). Patients with CR-POPF had increased levels of preoperative creatinine (CRE) and POD1 CRE. The combination of POD1 CRE (>104 μmol/L or not) and POD1 DFA (>5,000 U/L or not) stratified patients into subgroups with the maximum difference in CR-POPF risk. The CR-POPF rates were 17.82% (36/202) in group A (POD1 CRE ≤104 μmol/L and POD1 DFA ≤5,000 U/L), 53.33% (8/15) in group B (POD1 CRE >104 μmol/L and POD1 DFA ≤5,000 U/L), and 77.46% (55/71) in group C (POD1 DFA >5,000 U/L). The PPH rates were 1.98% (4/202), 20.00% (3/15), and 19.72% (14/71) in groups A, B, and C, respectively.

CONCLUSION

Patients with POD1 DFA ≤5,000 U/L and POD1 CRE >104 μmol/L have a high risk of CR-POPF and may not benefit from early drain removal. Patients with POD1 DFA ≤5,000 U/L and POD1 CRE ≤104 μmol/L have low risk of CR-POPF and PPH.

Development and validation of a network calculator model for safety and efficacy after pancreaticoduodenectomy in the elderly patients with pancreatic head cancer

BACKGROUND

Benefiting from increased life expectancy and improved perioperative management, more elderly patients with pancreatic head cancer (PHC) underwent pancreaticoduodenectomy (PD). However, individualized predictive models for the safety and efficacy of PD is still lacking. this study aimed to developed three safety- and efficacy-related risk calculators for elderly (> = 65 years) PHC patients.

METHODS

This study was designed with two research cohorts, namely, the training cohort and the validation cohort, and comprises four general steps: (1) Risk factors were analyzed for the incidence of postoperative complications, cancer-specific survival (CSS), and overall survival (OS) in the training cohort (N = 271) using logistic and Cox-regression analysis. (2) Nomograms were then plotted based on the above results. (3) The accuracy of the developed nomogram models was then verified with the validation cohort (N = 134) data using consistency index (C-index) and calibration curves. (4) We then evaluated the efficacy of these nomograms using decision curve analysis (DCA) in both the training and validation cohorts, and ultimately constructed three online calculators based on these nomograms.

RESULTS

We identified ASA, diabetes, smoking, and lymph node invasion as predisposing risk factors for postoperative complications, and the predictive factors that affected both OS and CSS were ASA, diabetes, BMI, CA19-9 level, and tumor diameter. By integrating the above risk factors, we constructed three nomograms on postoperative complication, CSS, and OS. The C-index for complication, CSS, and OS were 0.824, 0.784, and 0.801 in the training cohort and 0.746, 0.718, and 0.708 in the validation cohort. Moreover, the validation curves and DCA demonstrated good calibration and robust compliance in both training and validation cohorts. We then developed three web calculators (https://caiming.shinyapps.io/CMCD/, https://caiming.shinyapps.io/CMCSS/, and https://caiming.shinyapps.io/CMOS/) to facilitate the use of the nomograms.

CONCLUSIONS

The calculators demonstrated promising performance as an tool for predicting the safety and efficacy of PD in elderly PHC patients.

Serum amylase on postoperative day 1 is superior to serum lipase in predicting clinically relevant pancreatic fistula after partial pancreaticoduodenectomy

PURPOSE

To compare the predictive value of serum amylase and lipase regarding the occurrence of clinically relevant postoperative pancreatic fistula (cr-POPF) after partial pancreaticoduodenectomy (PD).

METHODS

Data from 228 consecutive patients undergoing PD were obtained from a prospective database. Serum amylase and lipase were measured on postoperative days (PODs) 0-2. Receiver-operating characteristics analysis was performed and cutoff values were tested using logistic regression.

RESULTS

Serum amylase had a larger area under the curve (AUC) on POD1 (AUC 0.89, p <0.001) than serum lipase. For serum amylase POD 1, a cutoff value of 70 U/l showed sensitivity and specificity of 100% and 70% for the diagnosis of cr-POPF. Serum amylase POD 1 > 70 U/l (OR 9.815, 95% CI 3.683-26.152, p < 0.001), drain amylase POD 1 > 300 U/l (OR 2.777, 95% CI 1.071-7.197, p= 0.036), and a small (≤ 3mm) pancreatic duct diameter (OR 3.705, 95% CI 1.426-9.627, p= 0.007) were significant predictors of cr-POPF in the multivariable analysis. Patients were divided into three risk groups based on serum amylase POD 1 and pancreatic duct diameter. This model had a good performance in discriminating cr-POPF (AUC 0.846, 95% CI 0.793-0.898). The sensitivity, specificity, and negative predictive value for the combination of serum amylase POD 1 <70 U/l and pancreatic duct diameter >3 mm were 100%, 70%, and 100%.

CONCLUSION

Serum amylase POD 1 was superior to serum lipase in predicting cr-POPF after PD. The proposed risk prediction model had a sensitivity and negative predictive value of 100%, allowing for early identification of cr-POPF.

A Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis analysis to evaluate the quality of reporting of postoperative pancreatic fistula prediction models after pancreatoduodenectomy: A systematic review

BACKGROUND

Postoperative pancreatic fistula is a frequent and potentially lethal complication after pancreatoduodenectomy. Several models have been developed to predict postoperative pancreatic fistula risk. This study was performed to evaluate the quality of reporting of postoperative pancreatic fistula prediction models after pancreatoduodenectomy using the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) checklist that provides guidelines on reporting prediction models to enhance transparency and to help in the decision-making regarding the implementation of the appropriate risk models into clinical practice.

METHODS

Studies that described prediction models to predict postoperative pancreatic fistula after pancreatoduodenectomy were searched according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The TRIPOD checklist was used to evaluate the adherence rate. The area under the curve and other performance measures were extracted if reported. A quadrant matrix chart is created to plot the area under the curve against TRIPOD adherence rate to find models with a combination of above-average TRIPOD adherence and area under the curve.

RESULTS

In total, 52 predictive models were included (23 development, 15 external validation, 4 incremental value, and 10 development and external validation). No risk model achieved 100% adherence to the TRIPOD. The mean adherence rate was 65%. Most authors failed to report on missing data and actions to blind assessment of predictors. Thirteen models had an above-average performance for TRIPOD checklist adherence and area under the curve.

CONCLUSION

Although the average TRIPOD adherence rate for postoperative pancreatic fistula models after pancreatoduodenectomy was 65%, higher compared to other published models, it does not meet TRIPOD standards for transparency. This study identified 13 models that performed above average in TRIPOD adherence and area under the curve, which could be the appropriate models to be used in clinical practice.

Postoperative pancreatitis and pancreatic fistulae: a review of current evidence

OBJECTIVES

Postoperative pancreatic fistula (POPF) represents one of the most severe complications following pancreatic surgery. Despite being a leading cause of morbidity and mortality, its pathophysiology is poorly understood. In recent years, there has been growing evidence to support the role of postoperative or post-pancreatectomy acute pancreatitis (PPAP) in the development of POPF. This article reviews the contemporary literature on POPF pathophysiology, risk factors, and prevention strategies.

METHODS

A literature search was conducted using electronic databases, including Ovid Medline, EMBASE, and Cochrane Library, to retrieve relevant literature published between 2005 and 2023. A narrative review was planned from the outset.

RESULTS

A total of 104 studies fulfilled criteria for inclusion. Forty-three studies reported on technical factors predisposing to POPF, including resection and reconstruction technique and adjuncts for anastomotic reinforcement. Thirty-four studies reported on POPF pathophysiology. There is compelling evidence to suggest that PPAP plays a critical role in the development of POPF. The acinar component of the remnant pancreas should be regarded as an intrinsic risk factor; meanwhile, operative stress, remnant hypoperfusion, and inflammation represent common mechanisms for acinar cell injury.

CONCLUSIONS

The evidence base for PPAP and POPF is evolving. Future POPF prevention strategies should look beyond anastomotic reinforcement and target underlying mechanisms of PPAP development.

Risk factor analysis and prediction of postoperative clinically relevant pancreatic fistula after distal pancreatectomy

OBJECTIVE

Postoperative pancreatic fistula (POPF) following distal pancreatectomy (DP) is a serious complication. In the present study, we aimed to identify the risk factors associated with clinically relevant postoperative pancreatic fistula (CR-POPF) and establish a nomogram model for predicting CR-POPF after DP.

METHODS

In total, 115 patients who underwent DP at the General Hospital of Northern Theater Command between January 2005 and December 2020 were retrospectively studied. Univariate and multivariable logistic regression analyses were used to identify the independent risk factors associated with CR-POPF. Then, a nomogram was formulated based on the results of multivariable logistic regression analysis. The predictive performance was evaluated with receiver operating characteristic (ROC) curves. Decision curve and clinical impact curve analyses were used to validate the clinical application value of the model.

RESULTS

The incidence of CR-POPF was 33.0% (38/115) in the present study. Multivariate logistic regression analysis identified the following variables as independent risk factors for POPF: body mass index (BMI) (OR 4.658, P = 0.004), preoperative albumin level (OR 7.934, P = 0.001), pancreatic thickness (OR 1.256, P = 0.003) and pancreatic texture (OR 3.143, P = 0.021). We created a nomogram by incorporating the above mentioned risk factors. The nomogram model showed better predictive value, with a concordance index of 0.842, sensitivity of 0.710, and specificity of 0.870 when compared to each risk factor. Decision curve and clinical impact curve analyses also indicated that the nomogram conferred a high clinical net benefit.

CONCLUSION

Our nomogram could accurately and objectively predict the risk of postoperative CR-POPF in individuals who underwent DP, which could help clinicians with early identification of patients who might develop CR-POPF and early development of a suitable fistula mitigation strategy and postoperative management.

Serum lipase on postoperative day one is a strong predictor of clinically relevant pancreatic fistula after pancreaticoduodenectomy: A retrospective cohort

BACKGROUND

Increased postoperative serum amylase has been recently reported to be associated with increased postoperative morbidity, but studies on postoperative serum lipase are limited. The aim of this study was to evaluate the value of postoperative serum lipase in predicting clinically relevant postoperative pancreatic fistula (CR-POPF) after pancreaticoduodenectomy (PD).

METHOD

A retrospective analysis was performed on 212 patients who underwent PD from September 2018 and March 2021, focusing on the association between postoperative day (POD) 1 serum lipase and CR-POPF.

RESULTS

Overall, 108 (50.9%) patients had elevated serum lipase levels (>68 U/L) on POD 1. Patients with elevated serum lipase exhibited a significantly higher incidence of CR-POPF (37.0% vs. 6.7%, p < 0.001). Receiver operating characteristic (ROC) analyses showed improved diagnostic accuracy for POD 1 serum lipase compared with POD 1 serum amylase in predicting CR-POPF (AUC: 0.801 vs. 0.745, p = 0.029). Elevated serum lipase on POD 1 and elevated serum CRP on POD 3 were identified as independent predictors of CR-POPF. A simple early postoperative model, consisting of POD 1 serum lipase levels and POD 3 serum CRP levels, showed good discrimination (AUC 0.76, 95% CI 0.69-0.83) to identify the onset of CR-POPF.

CONCLUSION

Serum lipase on POD 1 outperformed serum amylase on POD 1 in predicting CR-POPF after PD. The combination of POD 1 serum lipase and POD 3 serum CRP provides a reliable predicting model for CR-POPF.