Ommaya reservoir converted to permanent cerebrospinal fluid shunt in neonatal posthemorrhagic hydrocephalus: a risk factors analysis

Objective To identify factors influencing the transition to a permanent ventriculoperitoneal (VP) shunt in neonates with severe intraventricular hemorrhage (IVH) initially managed with an Ommaya reservoir. Methods This retrospective study involved 26 neonates diagnosed with IVH in Hebei Children's Hospital from January 2020 to January 2025. These patients were diagnosed with Papile grade III or IV IVH and treated with an Ommaya reservoir. They were categorized into two groups: a permanent shunt group ( n  = 10) and a non-permanent shunt group ( n  = 16). These two groups were compared based on demographic, clinical, and cerebrospinal fluid (CSF) variables. Results Of all study subjects, 10 patients (38.5%) received a permanent VP shunt placement. The 5-minute Apgar score was significantly higher in the permanent shunt group (9.40 ± 1.08) compared to that of the control group (7.25 ± 2.50, p  = 0.021). Also, the CSF protein levels were notably increased in the permanent shunt group (2.80 ± 1.43 g/L vs. 1.81 ± 0.79 g/L, p  = 0.028). And, ROC curve analysis revealed that a 5-minute Apgar score exceeding 8.5 was predictive of the requirement for permanent cerebrospinal fluid shunting, with corresponding sensitivity and specificity values of 80.0% and 68.7%, respectively. Moreover, multivariate analysis indicated that a 5-minute Apgar score greater than 8.5 remained an independent predictor of permanent shunt placement (adjusted OR 12.24, 95% CI 1.11–134.75, p  = 0.041). In contrast, CSF protein ≥2.04 g/L (adjusted OR 4.82, 95% CI 0.61–38.41, p  = 0.138) and sex did not show statistical significance. Conclusion In this retrospective study, a higher 5-minute Apgar score (>8.5) was identified as an independent predictor for the requirement of a permanent VP shunt after Ommaya reservoir placement in neonates with severe IVH. This seemingly counterintuitive result may be explained by the impact of a more robust neonatal circulation on the initial hemorrhage volume. Although elevated CSF protein was associated with shunt dependency, it was not an independent predictor. These findings underscore the intricate pathophysiological mechanisms of posthemorrhagic hydrocephalus.