OBJECTIVEIntra-ventricular hemorrhage (IVH) is one of the most serious complications of preterm infants. Significant numbers of the surviving infants with severe IVH go on to develop post-hemorrhagic hydrocephalus (PHH). The management of PHH remains a very challenging problem for both neonatologists and pediatric neurosurgeons. This study aimed to evaluate the efficacy and safety of the use of Ommaya reservoirs and serial cerebrospinal fluid (CSF) drainage in the management of a series of neonates with PHH.

METHODBetween January 1, 2003 and December 30, 2005, 15 consecutive newborn infants with IVH grades III to IV, complicated with progressive ventricular dilatation, underwent placement of an Ommaya reservoir. CSF was intermittently aspirated percutaneously from the reservoir. The amount and frequency of CSF aspiration were based on the clinical presentation and the follow-up results of serial cranial ultrasonograms or CT scans. The changes of CSF cell counts and chemistries were also followed. Patients whose progressive ventricular dilatation persisted despite serial CSF aspiration through Ommaya reservoir eventually had ventriculo-peritoneal shunts (V-P shunt) placed. All the patients were followed up in the outpatient clinic after discharge from the hospital and the neurodevelopmental outcomes were evaluated through 18-36 months of age.

RESULTA total of 15 infants were included in this series. Of them, 11 were preterm infants who were at gestational ages of 29 to 34 weeks and 4 infants were full-term. All of the 4 full term infants presented with progressive ventricular dilatation after suffering from the intra-cranial hemorrhage (3 infants were due to vitamin K deficiency and 1 was due to birth trauma). Thirteen infants had grade III IVH, and 2 had grade IV IVH based on initial cranial ultrasonographic and CT scans. The mean age when IVH was diagnosed was (9 +/- 1) days in preterm infants and (22 +/- 7) days in full-term infants; the mean age when Ommaya reservoir was placed was (18 +/- 11) days in preterm infants and (31 +/- 7) days in full-term infants. All the infants tolerated the surgical procedure well. The Ommaya reservoir was tapped for an average of (21.5 +/- 4.6) times per patient. The mean CSF volume per tap was (10.2 +/- 1.3) ml/kg. The values of CSF protein, glucose and cell counts slowly reached normal levels at approximately 3 - 5 weeks after the placement of the reservoir. The velocity of head circumference increase per week was less than 1 cm in 13 patients in 1 - 4 weeks after the placement of the reservoir and the size of ventricles decreased gradually. By 12 - 18 months, 12 infants had normal size ventricles, and 1 patient still had mild ventricular dilation at 36 months. Two infants developed progressive hydrocephalus after serial CSF aspiration through Ommaya reservoir. One infant had a V-P shunt placed at 2 months of age and another infant died at 3 months of age at home after parents refused further therapy. Complications consisted of reservoir leaking and CSF infection at 16th day of placement in one patient after repeated tapping. By the end of 18 - 36 months of follow-up, 11 of 14 infants were considered normal, two patients had mild impairment in neurodevelopmental outcome (both had spastic bilateral lower limbs paresis, and one of whom also had amblyopia) and the other had seizure disorder.

CONCLUSIONThe results from this series indicate that the placement of an Ommaya reservoir is relatively safe in newborn infants and is useful in the initial management of neonates with PHH and may be beneficial in improving their neurodevelopmental outcomes. A multicenter randomized trial may be needed to further validate the results of this report.