MATERNAL, PLACENTAL, AND NEONATAL ASSOCIATIONS WITH EARLY GERMINAL MATRIX INTRAVENTRICULAR HEMORRHAGE IN INFANTS BORN BEFORE 32 WEEKS GESTATION

This study tests the hypothesis that histologic placental lesions were significantly related to incidence of early or late germinal matrix/i ntraventricular hemorrhage (GM/IVH) in infants of less than 32 weeks' gestation independent of maternal or neonatal factors. Maternal and ne onatal charts of 406 singleton liveborn nonanomalous infants horn at l ess than 32 weeks' gestation were studied retrospectively for principa l indication for delivery, delivery mode, timing of antenatal steroid treatment, diagnosis of labor and augmentation, tocolysis, fetal prese ntation, and umbilical arterial and venous blood gas values. Extracted from neonatal charts were gestational age, growth measurements, initi al hematocrit and white blood cell count, administration of surfactant , and in the first 3 days of life, the use of presser agents and volum e expansion, lowest blood pressure, and data pertinent to respiratory function. Placental histologic examination was reviewed for various le sions, including histologic acute inflammation (graded on a scale of 0 to 4). GM/IVH (grades 1 to 4) diagnosed ultrasonographically less tha n 72 hours after birth was ''early.'' GM/IVH diagnosed after 72 hours of life was defined as ''late.'' Of the 406 patients, 44 (10.8%) had e arly GM/IVH; 21 (4.9%) had late GM/IVH. Stepwise logistic regression s elected five factors independently related to increased early GM/IVH r isk: Histologic acute inflammation (p <0.002); gestational age in days (p = 0.053); antenatal steroid treatment less than 48 hours before bi rth (p <0.035); volume expansion in the neonate (p <0.030), and magnes ium sulfate tocolysis (p <0.025). Stepwise regression analysis conside ring the grade of GM/IVH changed the order of variables, with gestatio nal age and use of presser therapy being more strongly related to high er grade of GM/IVH than amnion inflammation. Delivery mode, presentati on, principal indication for delivery, presence/augmentation of labor, mean biophysical profile sco res, mean umbilical arterial and venous blood gas values, and surfactant therapy were not related to early GM/ IVH in univariate or multivariate analyses. Neonatal factors associate d (p <0.05) with amnion inflammation were volume expansion at delivery and in the first 3 days of life, low mean systolic pressure, low mean oxygen pressure, low initial hematocrit and cord pH, and increased in itial WBC and toxic granulations of neutrophils. Only gestational age, and no maternal or placental factors, was significantly related to la te GM/IVH. Infants who have placentas with acute amnion inflammation a nd receive volume expansion, born to mothers who receive less than 48 hour's exposure to antenatal steroids and are selected to receive magn esium sulfate tocolysis, have increased incidence of early but not lat e GM/IVH. Amnion inflammation is significantly