Teruo Kimura, Atsushi Nakagaki, Takahiro Sanada, Hirotaka Sato, Seiya Fujikawa, Yujiro Matsushima, Anna Araki, Shigeki Yamada, Kazuhiro Sako, Manabu Kinoshita
Long-term functional independence after shunt surgery for idiopathic normal pressure hydrocephalus varies widely, particularly in older patients with multiple comorbidities. We aimed to develop and prospectively validate a frailty-integrated prognostic prediction tool to support patient selection and preoperative decision-making. A retrospective cohort of 82 patients with idiopathic normal pressure hydrocephalus who underwent shunt surgery between 2005 and 2014 was used to develop a weighted idiopathic normal pressure hydrocephalus-specific frailty index based on multivariable logistic regression. The index incorporated neurological and non-neurological comorbidities, including dementia, cancer, depression, and movement disorders. The weighted idiopathic normal pressure hydrocephalus-specific frailty index was combined with preoperative modified Rankin Scale scores to estimate the probability of achieving functional independence (modified Rankin Scale ≤2) at 1, 2, and mid- to long-term follow-up (up to 4 years). Prospective validation was performed in an independent cohort of 30 patients treated between 2018 and 2023. The weighted idiopathic normal pressure hydrocephalus-specific frailty index demonstrated improved predictive performance, particularly for mid- to long-term outcomes (2-4 years), compared with conventional frailty indices, with higher correlation coefficients and lower Akaike Information Criterion values. Predicted probabilities of long-term independence declined stepwise with increasing frailty burden. In the validation cohort, predicted and observed outcomes showed good agreement, with adequate calibration confirmed by the Hosmer-Lemeshow test. This frailty-integrated prognostic prediction tool provides individualized estimates of long-term postoperative independence, particularly in the context of frailty and comorbidity burden. By incorporating comorbidity burden into preoperative assessment, the tool supports patient selection and shared decision-making and may facilitate realistic counseling and long-term care planning in older patients.