PREDICTION AND PREVENTION OF COMPLICATIONS IN PATIENTS WITH PRIMARY AND POSTOPERATIVE VENTRAL HERNIA UNDER CONDITIONS OF COMORBIDITY
Summary. Purpose of the study. Conduct a retrospective and prospective analysis of complications in patients operated on for primary and postoperative ventral hernia based on the use of multi-parameter neural network clustering.
Materials and research methods. The object of the study was 1419 patients who underwent surgical interventions on the anterior abdominal wall for primary ventral and postoperative ventral hernias.
Results. An analysis of cluster portraits during neural network clustering in 1419 patients with primary and postoperative ventral hernias revealed that multimorbidity, allogeneioplasty of the onlay type, especially when using heavy polypropylene nets, morbid obesity of the II-III degree, as well as the presence of undifferentiated dysplasia syndrome connective tissue creates the highest risk group for the development of local and general postoperative complications.
Conclusions. 1. The comorbidity index significantly increases the operational-anesthetic risk and affects the incidence of postoperative complications. 2. The introduction of a comprehensive program (Enhanced Recovery After Surgery) in the perioperative period, which we used in 28.96 % of patients with primary and postoperative ventral hernias under conditions of comorbidity, reduced the number of postoperative complications by 2.08 times and reduced the duration of inpatient treatment by 2-3 days.
2. Bishop C. M. Neural Networks for Pattern Recognition / Bishop C. M. - Oxford : Oxford University Press, 1995. - Р. 504.
3. Israelsson L. A. Prevention of incisional hernias: How to close a midline incision / Israelsson L. A., Millbourn D. // Surg Clin North Am. -2013. - No 93 (5). - Р. 1027-1040.
4. Helgstrand F. National results after ventral hernia repair / F. Helgstrand // Dan Med J. -2016. -No 63 (7). -Р. B5258.
5. Kaoutzanis C. Risk factors for postoperative wound infections and prolonged hospitalization after ventral/ incisional hernia repair / Kaoutzanis C., Leichtle S., Mouawad N., Welch K. B. et al // Hernia. - 2015. - No 19 (1). - Р. 113-123.
6. Martsenyuk V. On an indirect method of exponential estimation for a neural network model with discretely distributed delays / V. Martsenyuk // Electronic Journal of Qualitative Theory of Differential Equations. 2017. No 23. Р. 1-16.
7. Martsenyuk V. Indirect method of exponential convergence estimation for neural network with discrete and distributed delays / V. Martsenyuk // Electronic Journal of Qualitative Theory of Differential Equations. 2017. No 246. Р. 1-12.
8. Patel P. V. Ventral hernia repair in the morbidly obese patient: A review of medical and surgical approaches in the literature / Patel P. V., Merchant A. M. // Bariatr Surg Pract Patient Care. -2014. -No 9 (2). -Р. 61-65.
9. Skipworth J. A. Improved outcomes in the management of high-risk incisional hernias utilizing biologic mesh and soft-tissue reconstruction: a single center experience / Skipworth J. A., Vyas S., Uppal L., Floyd D., Shankar A. // World J Surg. - 2015. -No 38. -Р. 1026-1034.
10. Shepherd A. J. Second-Order Methods for Neural Networks : Fast and Reliable Training Methods for Multi-Layer Perceptrons / A. J. Shepherd. Shepherd. London : Springer, 1997. Р. 342.