Endoscopic third ventriculostomy in idiopathic normal pressure hydrocephalus

OBJECTIVE: To define the role and indications for an endoscopic third ventriculostomy (ETV) in patients with idiopathic normal pressure hydrocephalus (INPH). A series of 25 patients treated by endoscopic technique was analyzed, and the results were compared with those of 14 studies reporting patients treated by shunting. METHODS: Twenty-five patients with INPH were treated by ETV from January 1994 through December 2000. All were younger than 75 years of age, had a preoperative clinical history of 1 year or less, had prevalence of gait disturbance with scarce or mild dementia, had marked ventricular enlargement on magnetic resonance imaging (MRI), and had intracranial pressure values ranging from 8 to 12 mm Hg. All were studied by a phase-contrast MRI flow study 1 month after ETV. The 14 reviewed series of patients treated by shunting (all published after 1980) each include more than 25 patients, for a total of 777 patients. RESULTS: The overall rate of neurological improvement after ETV in our series was 72% (including two patients reoperated on because of absence of flow in the MRI scan); this percentage is slightly higher than that found in the 14 series of shunted patients (66%). Gait disturbance showed a high rate of improvement when compared with other symptoms, both in our ETV study and in other shunting series. Postoperative complications occurred only in one patient (4%) with an intracerebral frontal hemorrhage and in 37.9% of patients from the series including shunted patients. CONCLUSION: In patients with INPH showing short duration of symptoms, prevalence of gait disturbance, and slight mental impairment, ETV provides similar results to those of shunting. We suggest performing ETV in these patients and reserving shunting only for those who do not improve after ETV, despite the presence of cerebrospinal fluid flow through the ventriculostomy on MRI flow studies. The good results after ETV in our series indirectly confirm that the cerebrospinal fluid absorption is good or at least sufficient in selected patients with INPH.