gms | German Medical Science

Objective: Treatment protocols for ventricular hemorrhage in neonates vary significantly between centers. Neuroendoscopic lavage (NL) has been demonstrated to be feasible, safe and effective in recent publications. We have thus adopted this technique and describe our initial experience and learning curve.

Methods: We retrospectively analyzed our initialcase series of this new technique in neonatal neurosurgerywith regard to learning curve.

Results: Five neonates born at a mean gestational age of 35 weeks (range 27-40 weeks) were analyzed. Four patients had symptomatic PHH with a mean preoperative frontal to occipital horn ratio of 0.58 (range 0.42-0.69). One term newborn presented with acute thalamic hemorrhage 20 days after birth. NL was performed at a mean of 20 days after birth (range 7-39 days). Pre-NL CSF protein was 317mg/dl (range 130-600mg/dl). NL of both lateral ventricles and third ventricle was performed via a unilateral approach and septostomy. NL was combined with Ommaya reservoir (N=2), Ommaya reservoir and ETV (N=1) and VPS (N=2). After initial treatment with NL and reservoir, a VPS was implanted after 20 and 26 days, respectively. NL reduced mean CSF protein to 60mg/dl (range 15-104mg/dl) and sonographic blood clot volume. In shunted infants (N=4), no VPS occlusion was observed within 30 days. No direct complications of NL occurred, but one CSF leak required revision.

Conclusion: NL reduces CSF protein and blood clot volume. No complications or early VPS failures were encountered in this small initial series, but certainly larger cohorts are required to assess the value of NL with regard to improving VPS outcome or even avoiding VPS in PHH. In our subjective learning curve, prior pediatric neuroendoscopic experience, interdisciplinary agreement on treatment, initial patient selection of term newborns and adherence to the published protocol allow safe implementation of this promising technique.