On the method of a randomised comparison of programmable valves with and without gravitational units: the SVASONA study.

OBJECTIVE: The supremacy of low-pressure valves (LPV) in the therapy of patients with idiopathic normal pressure hydrocephalus (iNPH) has been proven by the Dutch NPH study. The downside of LPVs is the high rate of overdrainage complications. In the meantime gravitational units have been developed with the objective of minimising overdrainage complications. Do these gravitational units allow the same favourable outcomes as in the Dutch NPH study without overdrainage complications? The goal of this prospective randomised controlled multicentre trial is to compare the rate of overdrainage complications after shunt surgery with programmable valves with or without a gravitational unit. METHOD: Patients suspected with iNPH with gait ataxia and one or two other symptoms of Hakim's triad and enlarged ventricles on CT or MRI underwent a cerebrospinal fluid (CSF) infusion test, CSF tap test and/or continuous ICP measurement in seven experienced centres. If iNPH was diagnosed the patients were randomised to receive a ventriculoperitoneal shunt with a programmable valve with or without a gravitational unit. The patients were followed up for 3, 6 and 12 months after surgery. The primary hypothesis of the study was that programmable valves with gravitational units will reduce the rate of overdrainage from 25% to 10%. The outcome measured by iNPH-specific outcome scales (Kiefer score, Black grading scale) and the generic quality of life (short form 12, SF12) was defined as the secondary endpoint. To detect the hypothesised difference in the incidence of overdrainage with a type I error of 5% and a type II error of 20%, correcting for multiple testing and an anticipated drop-out rate of 10%, we planned to enrol 250 patients. To ensure safety and efficacy we performed a planned interim analysis halfway. RESULTS: The study design has been found to be effective with regard to the primary hypothesis. CONCLUSION: The design of the SVASONA study was developed to be able to confirm the primary hypothesis. Thus, the method of the study should solve the dilemma of the Dutch NPH study by the randomised comparison of LPVs with and without gravitational units.