MATERIAL and METHODS: This retrospective study analyzed patients admitted to the neonatal intensive care unit of a Bahcesehir University School of Medicine-Affiliated Hospital due to preterm birth or low birth weight between June 1, 2012, and April 30, 2021. Categorical values were evaluated by Pearson chi-square or Fisher?s exact test. The Mann-Whitney U test compared continuous values between the groups. Logistic regression was used to evaluate the factors that affected permanent cerebrospinal fluid (CSF) diversion.

RESULTS: The study finally evaluated 180 newborns. Ninety-one newborns (50.5%) had grade I, 18 (10%) had grade II, 22 (12.2%) had grade III, and 49 (27.2%) had grade IV hemorrhage. One hundred and forty-nine patients (82.8%) were delivered by cesarean section, and 31 (17.2%) were delivered vaginally. All patients with low-grade hemorrhage who needed temporary CSF diversion eventually required permanent CSF diversion. For high-grade hemorrhage, 15 (grade III, 1; grade IV, 14) of 51 (29.4%) patients with ventricular access device (VAD) insertion required permanent CSF diversion. Fifteen (grade III, 6; grade IV, 9) of these 51 (29.4%) patients did not need permanent CSF diversion; thus, their VADs were removed.

CONCLUSION: The permanent CSF diversion rate was significantly higher in the high-grade hemorrhage group, which had significantly lower weight and gestational age at birth. Moreover, only weight at VAD insertion had minimal effect on the need for permanent CSF diversion. Keywords : Germinal matrix hemorrhage, Posthemorrhagic hydrocephalus, Ventricular access device, Permanent CSF diversion