The association between neonatal intracranial hemorrhage and hypernatremia has been reported. 1 Transient hypernatremia has been noted in newborn infants following exchange transfusionY An increased incidence of intracran]al hemorrhage in association with exchange transfusion, however, has not been reported. The present communication suggests a relationship of intracranial hemorrhage, exchange transfusion, and hypernatremia. MATERIAL AND METHODS Thirty-two infants included in this retrospective study represented all.of the neonatal deaths over a 2~A-year period in our hospital.All were infants who survived eight or more hours, who had at least one serum sodium determination recorded, and on whom an autopsy was performed. Hypernatremia was defined as a sodium concentration of 150 mEq per liter or more. Exchange transfusion was performed with either packed ACD or CPD blood, less than five days old, reconstituted with freshly thawed plasma. Sodium bicarbonate was neither infused during the exchahge nor added to the donor blood. RESULTS The study population was divided into three groups: Group I, seven infants, had no intracranial hemorrhage; Group II, 19 infants, had germinal plate, subependymal, or intraventricular hemorrhage; and Group III, six infants, had subarachnoid, subdural, or cortical hemorrhage. The mean Apgar scores at five minutes for all groups were similar. The proportion of very low birth weight (VLBW) infants (< 1,501 gm) was similar in Groups II and III, but significantly less in Group I (P < 0.025). All infants in Groups II and III required mechanical ventilation �9 for respiratory distress, except �9 for two infants in Group II. The frequencies of exchange transfusion in the three groups were significantly different (P < 0.025), since 11 of the 12 trans'fused infants were from Group II. Hypernatremia was documented in. 14 infants: 13 in Group II, one in Group I, and n0ne:'in Group III. These differences were highly significant (P < 0.005). The incidence of hypernatremia in the VLBW infants in contrast with that in larger infants was significantly greater (P < 0.05) (Table 1). Ten of the 13 VLBW infants in Group II had exchange transfusions. Six of these ten infants were normonatremic and four Werehypernatremic prior to exchange transfusion. Nine of the infants either remained hypernatremic or became hypernatremic after exchange transfusion. Only one of the five VLBW infants in Group II! had an exchange transfusion and that infant remained normonatremic.
Read full abstract