Correlation of ultrasound with postmortem neuropathologic studies in neonates

• Published in: Pediatric neurology Vol. 19; no. 4; pp. 263 - 271
• Main Authors: Adcock, Lisa M, Moore, Patricia J, Schlesinger, Alan E, Armstrong, Dawna L
• Format: Journal Article
• Language: English
• Published: New York, NY Elsevier Inc 01-10-1998; Elsevier
• Subjects: Biological and medical sciences; Brain - abnormalities; Brain - pathology; Brain Ischemia - diagnostic imaging; Brain Ischemia - pathology; Brain Neoplasms - diagnostic imaging; Brain Neoplasms - pathology; Cadaver; Cerebral Hemorrhage - diagnostic imaging; Cerebral Hemorrhage - pathology; Echoencephalography - standards; Female; Humans; Hydrocephalus - diagnostic imaging; Hydrocephalus - pathology; Infant, Newborn; Intensive Care Units, Neonatal; Male; Medical sciences; Nervous system (semeiology, syndromes); Nervous system as a whole; Neurology; Retrospective Studies; Sonography; Human; Transcranial route; Pathology; Nervous system diseases; Newborn; Echography; Postmortem; Exploration; Comparative study
• ISSN: 0887-8994; 1873-5150
• DOI: 10.1016/S0887-8994(98)00041-1

We examined the correlation between cranial sonogram and postmortem examination neurodiagnoses in 51 infants (30 boys and 21 girls) who underwent cranial sonogram before death. The mean gestation of the infants at birth was 32 weeks 6 days; the mean birth weight, 1,992 gm; and the mean survival, 27 days. Most infants had several postmortem examination neurodiagnoses; therefore the postmortem examination diagnosis considered to be most significant in terms of clinical management and long-term neurologic function was chosen for each infant and designated as the primary diagnosis. The ability of the cranial sonogram to diagnose the primary diagnosis was then evaluated. The accuracy of the cranial sonogram in defining primary diagnoses was 59%. There were 21 cases (41%) in which cranial sonogram failed to define the primary diagnosis; nine of these could be explained on the basis of the timing of the ultrasound in relation to the age of the lesion or to the microscopic nature of the lesion. In 12 patients the factors causing the failure of cranial sonogram to define primary postmortem examination diagnoses could not be identified. The positive predictive value of a cranial sonogram diagnosis was 77%; the negative predictive value was 19%.