Diseño y validación de un modelo predictivo de hemorragia intraventricular en neonatos pretérmino

• Autor: Quisse Sánchez, Carlos Eduardo
• Formato: tesis de grado
• Fecha de Publicación: 2020
• Institución: Universidad Nacional de Trujillo
• Repositorio: UNITRU-Tesis
• Lenguaje: español
• OAI Identifier: oai:dspace.unitru.edu.pe:20.500.14414/15573
• Enlace del recurso: https://hdl.handle.net/20.500.14414/15573
• Nivel de acceso: acceso abierto
• Materia: Hemorragia intraventricular; Recién nacidos; Pretérmino; Modelo predictivo; Sistema de puntuación de riesgo

OBJECTIVES. To design and validate a clinical predictive model of intraventric-ular hemorrhage (IVH). DESIGN. Retrospective nested case-control study. PAR-TICIPANTS. Preterm newborns (PTNBs) hospitalized in NICU from 2010 to 2017. METHODS. Predictive IVH models were designed in PTNBs from the "design" subcohort (DSC) (nDSC =220) using binary logistic regression. The obtained models were compared each other and the one with the best predictive perfor-mance (discrimination and calibration) was selected to be applied and assess-ment in the "validation" subcohort (VSC) (nVSC =110). RESULTS. Five predictors, including chorioamnionitis, gestational age <28 weeks, birth weight <1200 g, cardiopulmonary resuscitation and need for mechanical ventilatory support, were associated significantly with having IVH in PTNBs and were included in the DSC-derived predictive model. The coefficients for the DSC-derived and the VSC-derived predictive models were similar (p-value <0.05). The DSC-derived predictive model had good discrimination (area under the ROC curve: 0.79) and calibration (Hosmer-Lemeshow goodness-of-fit test, p-value >0.05) when was applied to the VSC. CONCLUSIONS. The DSC-derived predictive model had a suitable predictive performance in the VSC. A scoring system was developed to provide a simplified and categorized estimate of IVH risk in PTNBs.