Brain changes on magnetic resonance imaging in school-age children who had been preterm infants with intracranial hemorrhage / Alterações cerebrais na ressonância magnética em escolares que nasceram prematuramente e tiveram hemorragia intracraniana

ABSTRACT

Abstract Objective: To determine whether preterm infants diagnosed with intracranial hemorrhage (by transfontanellar ultrasound) at birth have cerebral lesions that are detectable by magnetic resonance imaging (MRI) upon reaching school age. Materials and Methods: MRI scans of the brain were obtained in 22 school-age children. Fifteen had presented intracranial hemorrhage at birth, and seven had not. We calculated the odds ratio (OR) for the occurrence of brain alterations detectable by MRI and the kappa index for discrepancies among the radiological reports. Results: The children without previous intracranial hemorrhage presented normal MRI findings. Of the 15 children with previous intracranial hemorrhage, 6 (40%) presented brain alterations on MRI: isolated ventricular alteration (dilation and asymmetry), in 2 (13.3%); and ventricular asymmetry accompanied by parenchymal lesion, in 4 (26.7%). The nine remaining children with previous intracranial hemorrhage (60%) presented normal MRI findings. The children with previous intracranial hemorrhage were more likely to present ventricular alteration (OR = 7.8) and parenchymal lesions (OR = 5.4). Conclusion: Ventricular and parenchymal brain alterations detected by MRI suggest isolated morphologic alterations that do not result in neurological impairment detectable on physical examination in school-age children.

RESUMO

Resumo Objetivo: Investigar se ocorrem lesões cerebrais detectáveis na ressonância magnética (RM) cerebral em escolares que foram prematuros e tiveram hemorragia intracraniana diagnosticada pela ultrassonografia transfontanelar ao nascer. Materiais e Métodos: Foi realizada RM cerebral em 22 escolares, sendo 15 com história de hemorragia intracraniana e 7 não. Calculou-se a odds ratio (OR) de ocorrer alterações cerebrais detectáveis na RM cerebral e o valor de kappa para avaliar discrepâncias entre laudos radiológicos. Resultados: Os escolares sem hemorragia intracraniana apresentaram RM cerebral normal. Dos 15 pacientes com hemorragia intracraniana, 6 (40%) tiveram alteração na RM cerebral na idade escolar, sendo 2 (13,3%) com alteração ventricular isolada e 4 (26,7%) com assimetria ventricular associada a lesão parenquimatosa. Nove escolares (60%) com hemorragia intracraniana apresentaram RM cerebral normal. Os escolares com hemorragia intracraniana tiveram maior chance de apresentar alteração ventricular (OR = 7,8) e lesão parenquimatosa (OR = 5,4) na RM cerebral. Conclusão: As alterações ventriculares e parenquimatosas detectadas na RM cerebral sugerem alterações morfológicas isoladas, que não constituem comprometimento neurológico detectável ao exame físico na idade escolar.