Abstract
BACKGROUND: Patients with idiopathic normal-pressure hydrocephalus (iNPH) are predisposed to developing dementing disorders. Cerebrospinal fluid (CSF) shunt implantation is a treatment used to improve the motor and cognitive disabilities of these patients; however, its effect on the risk of developing dementing disorders remains unclear. We conducted a population-based propensity-weighted cohort study to investigate whether CSF shunt surgery may reduce the risk of subsequently developing dementia, Alzheimer's disease (AD), and vascular dementia in iNPH patients. METHODS: Patients aged ≥ 60 years who were diagnosed with iNPH (n = 2053) between January 2001 and June 2018 were identified from the Taiwan National Health Insurance Research Database. Various demographic characteristics (age, sex, and monthly income) and clinical data (incidence year, comorbidities, and Charlson comorbidity index) were collected and divided into the shunt surgery group (SSG) and the non-shunt surgery group (NSSG). Stabilized inverse probability of treatment weighting by using the propensity score was performed to achieve a balanced distribution of confounders across the two study groups. The cumulative incidence rate and risk of dementing disorders were estimated during a 16-year follow-up period. RESULTS: After weighting, the data of 375.0 patients in SSG and 1677.4 patients in NSSG were analyzed. Kaplan-Meier curve analysis indicated that the cumulative incidence rate of AD (p = 0.009), but not dementia (p = 0.241) and vascular dementia (p = 0.761), in SSG was significantly lower than that in NSSG over the 16-year follow-up period. Cox proportional hazards regression analysis revealed that SSG had a reduced hazard ratio (HR) for developing AD [HR (95% CI) 0.17 (0.04-0.69)], but not for dementia [HR (95% CI) 0.83 (0.61-1.12)] and vascular dementia [HR (95% CI) 1.18 (0.44-3.16)], compared with NSSG. Further Fine-Gray hazard regression analysis with death as a competing event demonstrated that SSG had a reduced subdistribution HR (sHR) for developing dementia [sHR (95% CI) 0.74 (0.55-0.99)] and AD [sHR (95% CI) 0.15 (0.04-0.61)], but not for vascular dementia [sHR (95% CI) 1.07 (0.40-2.86)]. CONCLUSION: CSF shunt surgery is associated with reduced risks of the subsequent development of dementia and AD in iNPH patients. Our findings may provide valuable information for assessing the benefit-to-risk profile of CSF shunt surgery.