Objective：Currently, there is no reliable evidence on the usefulness of rehabilitation for idiopathic, normal-pressure hydrocephalus (iNPH）patients after cerebrospinal shunt surgery. Therefore, in this study we investigated the association between the duration of rehabilitation and outcomes.Methods：We evaluated the changes before and after cerebrospinal shunt surgery and rehabilitation by applying the 3-m timed up-and-go test (TUG), 10-m straight walk test, 180-degree turn, 30-seconds chair-stand test (CS-30), mini-mental state examination (MMSE), and frontal assessment battery (FAB) in 81 patients with iNPH. Clinical outcomes were classified as excellent, good, or unsatisfactory.Results：At discharge, TUG, 180-degree turn, and CS-30 were significantly improved in 38 patients who had been in rehabilitation for ≥2 weeks, compared to those with ＜2 weeks-rehabilitation after shunt surgery. However, there was no significant difference between MMSE and FAB tests in the two groups. Rehabilitation for ≥2 weeks significantly improved the patient outcome, especially for those patients with severe gait disturbance before treatment. In 53 patients whose initial TUG time was ≥13.5 s, rehabilitation for ≥2 weeks was effective, and resulted in an excellent outcome (odds ratio：4.52, 95％ confidence interval：1.22-18.7, P value：0.012).Conclusion：In-hospital rehabilitation after cerebrospinal shunt surgery was useful for patients with iNPH and severe disability in daily activities. Therefore, we suggest that these patients require in-hospital rehabilitation after shunt surgery until sufficient improvement of gait disturbance is achieved, in order to maximize activities of daily living.

Objective：Currently, there is no reliable evidence on the usefulness of rehabilitation for idiopathic, normal-pressure hydrocephalus (iNPH）patients after cerebrospinal shunt surgery. Therefore, in this study we investigated the association between the duration of rehabilitation and outcomes.Methods：We evaluated the changes before and after cerebrospinal shunt surgery and rehabilitation by applying the 3-m timed up-and-go test (TUG), 10-m straight walk test, 180-degree turn, 30-seconds chair-stand test (CS-30), mini-mental state examination (MMSE), and frontal assessment battery (FAB) in 81 patients with iNPH. Clinical outcomes were classified as excellent, good, or unsatisfactory.Results：At discharge, TUG, 180-degree turn, and CS-30 were significantly improved in 38 patients who had been in rehabilitation for ≥2 weeks, compared to those with ＜2 weeks-rehabilitation after shunt surgery. However, there was no significant difference between MMSE and FAB tests in the two groups. Rehabilitation for ≥2 weeks significantly improved the patient outcome, especially for those patients with severe gait disturbance before treatment. In 53 patients whose initial TUG time was ≥13.5 s, rehabilitation for ≥2 weeks was effective, and resulted in an excellent outcome (odds ratio：4.52, 95％ confidence interval：1.22-18.7, P value：0.012).Conclusion：In-hospital rehabilitation after cerebrospinal shunt surgery was useful for patients with iNPH and severe disability in daily activities. Therefore, we suggest that these patients require in-hospital rehabilitation after shunt surgery until sufficient improvement of gait disturbance is achieved, in order to maximize activities of daily living.