J Neurosurg Pediatr. 2026 Jun 19:1-10. doi: 10.3171/2026.1.PEDS25486. Online ahead of print.
ABSTRACT
OBJECTIVE: The aim of this study was to evaluate the clinical features and surgical complication rates of preterm infants with posthemorrhagic ventricular dilatation (PHVD) who underwent CSF diversion, comparing early intervention (EI) and late intervention (LI).
METHODS: This was a retrospective international multicenter study of infants born at ≤ 34 weeks’ gestational age (GA) and treated between 2018 and 2022 for grade III or IV germinal matrix hemorrhage and intraventricular hemorrhage, who required intervention for PHVD. The primary outcome of interest was the rate of postoperative surgical complications. Secondary outcomes were rate of temporizing neurosurgical procedures (TNPs) after initial lumbar puncture (LP) if performed, rate of conversion from TNP to a ventriculoperitoneal shunt (VPS), and overall rate of VPS insertion. Summary statistics and univariable and multivariable logistic regression were performed to determine variables predictive of complications.
RESULTS: One hundred seventy infants from 6 centers in the United States, Canada, and the Netherlands were included. Infants at LI centers were more premature (GA 26.7 ± 2.8 vs 29.3 ± 2.4 weeks, p < 0.001) with lower birth weights (BWs; 1004 ± 448 vs 1438 ± 462 g, p < 0.001) compared to those at EI centers. The first neurosurgical intervention occurred at a median day of life after birth (DOL) of 31.5 (IQR 24-45) days at LI centers versus 18 (IQR 13-21) days at EI centers (p < 0.001). Ventricular index (> 97th percentile) at the first neurosurgical intervention (TNP and/or VPS placement) was lower at EI centers (5.05 ± 2.32 vs 12.88 ± 5.70 mm, p < 0.001). Infant weight at the first neurosurgical intervention did not differ between site types (p = 0.466). EI centers had a larger proportion of patients who underwent LP (97% vs 58%, p < 0.001) and smaller proportion who required a VPS after TNP (42% vs 70%, p = 0.002). The overall VPS insertion rate was higher at LI centers (62% vs 30%, p < 0.001). There was no statistical difference in the complication rate between EI and LI centers (11% vs 24%, respectively, OR 0.42, 95% CI 0.12-1.24, p = 0.115). On univariable analysis, lower GA (p = 0.002), lower BW (p = 0.003), later DOL at first neurosurgical intervention (p = 0.035), diagnosis of meningitis before neurosurgical intervention (p = 0.047), and of necrotizing enterocolitis (p = 0.017) during the neonatal intensive care unit admission were predictive of complications. However, only lower GA (p = 0.029) and BW (p = 0.031) remained significant on multivariable analysis.
CONCLUSIONS: The neurosurgical complication rate did not differ between EI and LI centers. On multivariable analysis, neurosurgical complications were associated with younger GA and lower BW but not with variables regarding the timing of intervention. These observations support cautious early CSF diversion in more preterm infants with lower BWs.
PMID:42320053 | DOI:10.3171/2026.1.PEDS25486
