gms | German Medical Science

Objective: The etiology of adult hydrocephalus is heterogeneous with idiopathic NPH (50%), posthemorrhagic (23%) and posttraumatic hydrocephalus (12.5%) as most common. Mechanical failures of shunt systems are described with a vast range in 5% to 64%. Averaging the recent literature, valve-related under- and over-drainage syndromes occur in about 10-20% each and are associated with an increased morbidity. The valve selection is discussed controversial because of low level of evidence. Before May 2014 we implanted by default differential pressure valves without gravitational unit in all our patients. Afterwards, according to the SVASONA-Trial which showed that gravitational devices prevent over-drainage complications in idiopathic NPH patients, we changed our policy. Purpose of this study was to evaluate the implantation of differential pressure valves without gravitational unit concerning valve-related complications and the occurrence of those adverse events in relation to the hydrocephalus etiology.

Methods: We retrospectively analyzed all adult patients who underwent shunt implantation in our department between March 2010 and May 2014 concerning valve-related revisions. The relative risk of shunt failures was calculated by an univariate and multivariate regression model. Subgroups were characterized by the different hydrocephalus etiologies and alpha-level was determined on 5% to achieve statistical significance with a power of 0.9.

Results: Overall 510 patients were included and 273 (53.5%) were male. Median age was 69 (range 18 to 92) years. Valve-associated complications requiring revision surgery occurred in 93 (18.2%) patients, whereas 66 (12.9%) showed over-drainage with subdural hygroma and 27 (5.3%) mechanical valve dysfunction. Idiopathic NPH (n=137) and patients after tumor surgery (n=113) built the largest populations in addition to patients with hydrocephalus after trauma (n=58), intracerebral hematoma (n=80), subarachnoid bleeding (n=81) and other entities (n=41). Concerning these different etiologies, shunt implantation after tumor resection showed a significant lower risk for over-drainage complications (OR 0.4; CI 0.2-0.9; p<0.05). There was a trend to a reduced risk after subarachnoid bleeding (OR 0.5, CI 0.2-1.2, p=0.16) and increased rates in idiopathic NPH (OR 1.4; CI 0.8-2.5; p=0.20) and after intracerebral hematoma (OR 1.7; CI 0.8-3.1; p=0.12), without reaching significance. For valve under-drainage no significant differences for the several entities could be detected.

Conclusion: Hydrocephalus patients with differential pressure valves without gravitational units show no increased revision rates compared to the literature. The risk of over-drainage is reduced in patients receiving a shunt after intracranial tumor resection. Under-drainage rates do not differ significantly between the various hydrocephalus etiologies.