Prediction of pancreatic fistula after pancreatoduodenectomy using machine learning

Objective: to analyze the results of pancreatoduodenectomy (PD) and identify predictive risk factors for postoperative pancreatic fistula (PF) using machine learning (ML) technology.

Material and Methods. A nonrandomized study of treatment outcomes in 128 patients, who underwent PD for periampullary carcinoma between 2018 and 2023, was conducted. To predict PF, the ML models based on the multilayer perceptron and binary logistic regression (BLR) in SPSS Statistics v.26, were used. The Receiver Operator Characteristics (ROC) analysis was used to assess the accuracy of the models. To compare ROC curves, the DeLong test was used.

Results. Clinically significant PF occurred in 19 (14.8 %) patients (grade B according to ISGPS 2016 – in 16 (12.5 %), grade C – in 3 (2.3 %)). The data of 90 (70.3 %) patients were used to train the neural network, and 38 (29.7 %) were used to test the predictive model. In multivariate analysis, the predictors of PF were a comorbidity level above 7 points on the age-adjusted Charlson scale, a diameter of the main pancreatic duct less than 3 mm, and a soft pancreatic consistency. The diagnostic accuracy of the ML model estimated using the area under the ROC curve was 0.939 ± 0.027 (95 % CI: 0.859–0.998, sensitivity: 84.2 %, specificity; 96.3 %). The predictive model, which was developed using BLR, demonstrated lower accuracy: 0.918±0.039 (95 % CI: 0.842–0.994, sensitivity: 78.9 %, specificity: 94.5 %) (p=0.02).

Conclusion. The use of machine learning technologies makes it possible to increase the probability of a correct prediction of the occurrence of pancreatic fistula after pancreatoduodenectomy.

1. Solodky V.A., Kriger A.G., Gorin D.S., Dvukhzhilov M.V., Akhaladze G.G., Goncharov S.V., Panteleev V.I., Shuinova E.A. Pancreaticoduodenectomy – results and prospects (two-center study). Pirogov Russian Journal of Surgery. 2023; (5): 13–21. (in Russian). doi: 10.17116/hirurgia202305113.

2. Khatkov I.E., Domrachev S.A., Tsvirkun V.V., Izrailov R.E., Vasnev O.S., Kulezneva Yu.V., Les’ko K.A., Schadrova V.V., Nikitin B.S., Starostina N.S., Tyutyunnik P.S., Baychorov M.E., Andrianov A.V., Mikhnevich M.V. Prediction of postpancreatoduodenectomy pancreatic fistula with the use of computer tomography. Medical Visualization. 2019; (1): 19–27. (in Russian). doi: 10.24835/1607-0763-2019-1-19-27.

3. Egorov S.V., Petrov R.V. Simple and reliable pancreatoenteroanastomosis. Pirogov Russian Journal of Surgery. 2017; (11): 60–8. (in Russian). doi: 10.17116/hirurgia20171160-68.

4. Dalgatov K.D., Kurskov A.O., Khalbaginov A.A., Sazhin A.V. Pancreatodigestive anastomosis: from history to modernity. Pirogov Russian Journal of Surgery. 2021; (10): 81–6. (in Russian). doi: 10.17116/hirurgia202110181.

5. Singh G. Artificial intelligence in colorectal cancer: a review. Siberian Journal of Oncology. 2023; 22(3): 99–107. (in English). doi: 10.21294/1814-4861-2023-22-3-99-107.

6. Melnikov P.V., Dovedov V.N., Kanner D.Yu., Chernikovskiy I.L. Artificial Intelligence in surgical practice. Pelvic Surgery and Oncology. 2020; 10(3–4): 60–4. (In Russian). doi: 10.17650/2686-9594-2020-10-3-4-60-64.

7. Yin H., Zhang F., Yang X., Meng X., Miao Y., Noor Hussain M.S., Yang L., Li Z. Research trends of artificial intelligence in pancreatic cancer: a bibliometric analysis. Front Oncol. 2022; 12: 1–13. doi: 10.3389/fonc.2022.973999.

8. Dindo D., Demartines N., Clavien P.A. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004; 240(2): 205–13. doi: 10.1097/01.sla.0000133083.54934.ae.

9. Bassi C., Marchegiani G., Dervenis C., Sarr M., Abu Hilal M., Adham M., Allen P., Andersson R., Asbun H.J., Besselink M.G., Conlon K., Del Chiaro M., Falconi M., Fernandez-Cruz L., Fernandez-Del Castillo C., Fingerhut A., Friess H., Gouma D.J., Hackert T., Izbicki J., Lillemoe K.D., Neoptolemos J.P., Olah A., Schulick R., Shrikhande S.V., Takada T., Takaori K., Traverso W., Vollmer C.R., Wolfgang C.L., Yeo C.J., Salvia R., Buchler M.; International Study Group on Pancreatic Surgery (ISGPS). The 2016 update of the International Study Group (ISGPS) definition and grading of postoperative pancreatic fistula: 11 Years After. Surgery. 2017; 161(3): 584–91. doi: 10.1016/j.surg.2016.11.014.

10. Kit O.I., Frantsiyants E.M., Katelnitskaya O.V. Risk of venous thromboembolism in patients with pancreatic cancer. Siberian Journal of Oncology. 2022; 21(3): 24–32. (in Russian). doi: 10.21294/1814-4861-2022-21-3-24-32.

11. Kuchin D.M., Kolesnik Y.I., Torgomyan H.G., Zagainov V.E. Factors affecting overall survival in ductal adenocarcinoma of the pancreatic head. Experience of one center. Malignant Tumours. 2021; 11(1): 20–8. (in Russian). doi: 10.18027/2224-5057-2021-11-1-20-28.

12. Nazarova D.V., Rasulov R.I., Zubrinsky K.G., Songolov G.I. Evolution of treatment of cancer of the major duodenal papilla. Siberian Journal of Oncology. 2021; 20(1): 141–8. (in Russian). doi: 10.21294/1814-4861-2021-20-1-141-148.

13. Bonsdorff A., Sallinen V. Prediction of postoperative pancreatic fistula and pancreatitis after pancreatoduodenectomy or distal pancreatectomy: A review. Scand J Surg. 2023; 112(2): 126–34. doi: 10.1177/14574969231167781.

14. Kabanov M.Yu., Glushkov N.I., Semencov K.V., Koshelev T.E., Savchenkov D.K., Sizonenko N.A., Goloshchapova I.M. Modern approaches to the prevention and treatment of postoperative complications in pancreatic head cancer. Bulletin of Pirogov National Medical & Surgical Center. 2023; 18(2): 128–33. (in Russian). doi: 10.25881/20728255_2023_18_2_128.

15. Stoop T.F., Bergquist E., Theijse R.T., Hempel S., van Dieren S., Sparrelid E., Distler M., Hackert T., Besselink M.G., Del Chiaro M., Ghorbani P.; Collaborators. Systematic Review and Meta-analysis of the Role of Total Pancreatectomy as an Alternative to Pancreatoduodenectomy in Patients at High Risk for Postoperative Pancreatic Fistula: Is it a Justifiable Indication? Ann Surg. 2023; 278(4): 702–11. doi: 10.1097/SLA.0000000000005895.

16. Yoon S.J., Kwon W., Lee O.J., Jung J.H., Shin Y.C., Lim C.S., Kim H., Jang J.Y., Shin S.H., Heo J.S., Han I.W. External validation of risk prediction platforms for pancreatic fistula after pancreatoduodenectomy using nomograms and artificial intelligence. Ann Surg Treat Res. 2022; 102(3): 147–52. doi:10.4174/astr.2022.102.3.147.

17. Kambakamba P., Mannil M., Herrera P.E., Müller P.C., Kuemmerli C., Linecker M., von Spiczak J., Hüllner M.W., Raptis D.A., Petrowsky H., Clavien P.A., Alkadhi H. The potential of machine learning to predict postoperative pancreatic fistula based on preoperative, non-contrastenhanced CT: A proof-of-principle study. Surgery. 2020; 167(2): 448–54. doi: 10.1016/j.surg.2019.09.019.

18. Han I.W., Cho K., Ryu Y., Shin S.H., Heo J.S., Choi D.W., Chung M.J., Kwon O.C., Cho B.H. Risk prediction platform for pancreatic fistula after pancreatoduodenectomy using artificial intelligence. World J Gastroenterol. 2020; 26(30): 4453–64. doi:10.3748/wjg.v26.i30.4453.

19. Mu W., Liu C., Gao F., Qi Y., Lu H., Liu Z., Zhang X., Cai X., Ji R.Y., Hou Y., Tian J., Shi Y. Prediction of clinically relevant Pancreaticoenteric Anastomotic Fistulas after Pancreatoduodenectomy using deep learning of Preoperative Computed Tomography. Theranostics. 2020; 10(21): 9779–88. doi: 10.7150/thno.49671.

20. Ingwersen E.W., Stam W.T., Meijs B.J.V., Roor J., Besselink M.G., Groot Koerkamp B., de Hingh I.H.J.T., van Santvoort H.C., Stommel M.W.J., Daams F.; Dutch Pancreatic Cancer Group. Machine learning versus logistic regression for the prediction of complications after pancreatoduodenectomy. Surgery. 2023; 174(3): 435–40. doi: 10.1016/j.surg.2023.03.012.