Graft Survival in Liver Transplantation: An Artificial Neuronal Network Assisted Analysis of the Importance of Comorbidities
Zalba Etayo, María Begoña (Universidad de Zaragoza) ; Marín Araiz, Lucía ; Montes Aranguren, María (Universidad de Zaragoza) ; Lorente Pérez, Sara (Universidad de Zaragoza) ; Palacios Gasos, Pilar (Universidad de Zaragoza) ; Pascual Bielsa, Ana ; Sánchez Donoso,Nuria ; Serrano Aullo, Trinidad (Universidad de Zaragoza) ; Araiz Burdio, Juan José (Universidad de Zaragoza)
Resumen: Objectives: Liver transplant represents a widespread therapeutic option for patients with end-stage liver failure. Up to now, most of the scores describing the probability of liver graft survival have shown poor predictive performance. With this in mind, the present study seeks to analyze the predictive value of recipient comorbidities on liver graft survival within the first year.
Materials and Methods: The study included prospectively collected data from patients who received a liver transplant at our center from 2010 to 2021. A
predictive model was then developed through an Artificial Neural Network that included the parameters associated with graft loss as identified by the Spanish Liver Transplant Registry report and comorbidities with prevalence >2% present in our study cohort.
Results: Most patients in our study were men (75.5%); mean age was 54.8 ± 9.6 years. The main cause of transplant was cirrhosis (86.7%), and 67.4% of patients had some associated comorbidities. Graft loss due to retransplant or death with dysfunction occurred in 14% of cases. Of all the variables analyzed, we found 3 comorbidities associated with graft loss (as shown by informative value and normalized informative value, respectively): antiplatelet and/or anticoagulants treatments (0.124 and 78.4%), previous immunosuppression (0.110 and 69.6%), and portal thrombosis (0.105 and 66.3%). Remarkably, our model showed a C statistic of 0.745 (95% CI, 0.692-0.798; asymptotic P < .001), which was higher than others found in previous studies.
Conclusions: Our model identified key parameters that may influence graft loss, including specific recipient comorbidities. The use of artificial intelligence methods could reveal connections that may be overlooked by conventional statistics.
Idioma: Inglés
DOI: 10.6002/ect.2022.0372
Año: 2023
Publicado en: Experimental and Clinical Transplantation 21, 4 (2023), 338-344
ISSN: 1304-0855
Factor impacto JCR: 0.7 (2023)
Categ. JCR: TRANSPLANTATION rank: 26 / 31 = 0.839 (2023) - Q4 - T3
Factor impacto CITESCORE: 1.4 - Transplantation (Q3)
Factor impacto SCIMAGO: 0.247 - Transplantation (Q3)
Tipo y forma: Artículo (PostPrint)
Área (Departamento): Area Medicina (Dpto. Medicina, Psiqu. y Derm.)
Área (Departamento): Área Cirugía (Dpto. Cirugía)
Debe reconocer adecuadamente la autoría, proporcionar un enlace a la licencia e indicar si se han realizado cambios. Puede hacerlo de cualquier manera razonable, pero no de una manera que sugiera que tiene el apoyo del licenciador o lo recibe por el uso que hace. No puede utilizar el material para una finalidad comercial. Si remezcla, transforma o crea a partir del material, no puede difundir el material modificado.
Exportado de SIDERAL (2024-11-22-11:59:17)
Visitas y descargas
Materials and Methods: The study included prospectively collected data from patients who received a liver transplant at our center from 2010 to 2021. A
predictive model was then developed through an Artificial Neural Network that included the parameters associated with graft loss as identified by the Spanish Liver Transplant Registry report and comorbidities with prevalence >2% present in our study cohort.
Results: Most patients in our study were men (75.5%); mean age was 54.8 ± 9.6 years. The main cause of transplant was cirrhosis (86.7%), and 67.4% of patients had some associated comorbidities. Graft loss due to retransplant or death with dysfunction occurred in 14% of cases. Of all the variables analyzed, we found 3 comorbidities associated with graft loss (as shown by informative value and normalized informative value, respectively): antiplatelet and/or anticoagulants treatments (0.124 and 78.4%), previous immunosuppression (0.110 and 69.6%), and portal thrombosis (0.105 and 66.3%). Remarkably, our model showed a C statistic of 0.745 (95% CI, 0.692-0.798; asymptotic P < .001), which was higher than others found in previous studies.
Conclusions: Our model identified key parameters that may influence graft loss, including specific recipient comorbidities. The use of artificial intelligence methods could reveal connections that may be overlooked by conventional statistics.
Idioma: Inglés
DOI: 10.6002/ect.2022.0372
Año: 2023
Publicado en: Experimental and Clinical Transplantation 21, 4 (2023), 338-344
ISSN: 1304-0855
Factor impacto JCR: 0.7 (2023)
Categ. JCR: TRANSPLANTATION rank: 26 / 31 = 0.839 (2023) - Q4 - T3
Factor impacto CITESCORE: 1.4 - Transplantation (Q3)
Factor impacto SCIMAGO: 0.247 - Transplantation (Q3)
Tipo y forma: Artículo (PostPrint)
Área (Departamento): Area Medicina (Dpto. Medicina, Psiqu. y Derm.)
Área (Departamento): Área Cirugía (Dpto. Cirugía)
Debe reconocer adecuadamente la autoría, proporcionar un enlace a la licencia e indicar si se han realizado cambios. Puede hacerlo de cualquier manera razonable, pero no de una manera que sugiera que tiene el apoyo del licenciador o lo recibe por el uso que hace. No puede utilizar el material para una finalidad comercial. Si remezcla, transforma o crea a partir del material, no puede difundir el material modificado.Exportado de SIDERAL (2024-11-22-11:59:17)
Enlace permanente:
Visitas y descargas
Este artículo se encuentra en las siguientes colecciones:
Artículos > Artículos por área > Medicina
Artículos > Artículos por área > Cirugía
Registro creado el 2023-12-19, última modificación el 2024-11-25