The aim of the study was to compare the characteristics of ventriculosubgaleal shunts during the clinical course of posthemorrhagic and postinfectious hydrocephalus in the neonatal period. The study comprised 102 premature babies in whom subgaleal shunt was consecutively inserted between 2006 and 2011. Seventy-two patients had posthemorrhagic hydrocephalus (mean gestational age 27.3 ± 2.1 weeks, mean birth weight 1,036.9 ± 327.7 g, mean age at insertion 51.4 ± 56.2 days) and 30 patients were operated postinfectiously (27.5 ± 2.2 weeks, 1,064.7 g ± 310.7 g, 115.9 ± 47.8 days). The mean survival of subgaleal shunts was 87.9 days for the posthemorrhagic group and 75.6 days for the postinfectious group. Only six infants (8.3 %) did not need ventriculoperitoneal shunts later, all posthemorrhagic. There were meaningful differences between two groups with regard to ventriculosubgaleal shunt-related infections (8.3 % in posthemorrhagic versus 20.0 % in postinfectious) and shunt revision rate (6.9 % in posthemorrhagic versus 13.3 % in postinfectious), but these were not statistically significant. The need of ventriculoscopic procedures was notably more frequent in postinfectious group (1.4 versus 23.3 %). In premature infants with ventriculomegaly, the subgaleal shunt is an effective temporary diversion tool. The complications were less with posthemorrhagic than with postinfectious hydrocephalus. With previous severe infections of prematures, the risk for complications regarding infection and obstruction will be 2.75 and 2.06 (odds ratios) times higher and more frequent need of ventriculoscopic procedures should be considered (odds ratio 21.6).