A statistical study of autopsied cases of neonatal intraventricular hemorrhage

We have performed a statistical study of 1218 autopsies of neonates to extract the factors related to fatal neonatal intraventricular hemorrhage (IVH). Prematurity indicated by a short gestational period or a low birth weight was the most significant factor for IVH. Many other factors considered to be related to IVH were also related closely to prematurity. Therefore, we pose the following question throughout the study: Did the factors independently relate to IVH without any confounding effects of prematurity? To obtain the answer, we tried a statistical adjustment for gestational age to eliminate the confounding effects of prematurity. After the adjustment was made, we concluded that the following 10 items, in order, were the actual factors related to fatal IVH: prematurity, subependymal hemorrhage, subarachnoidal hemorrhage, respiratory distress syndrome, hyaline membrane disease, respiratory acidosis, intracerebral hemorrhage, cyanosis, respirator care, and low Apgar score. There were two distinctive categories in the factors that were eliminated by the adjustment. One was a group that has no significance whatsoever in any subdivided gestational groups, such as meconium aspiration syndrome, due to purely the result of the confounding effect. The other was a group that showed its statistical significance only in a certain period of gestation, such as breech presentation.