The goal of this observational study is to describe changes in cognitive profile and frailty from pre shunt to one year after the shunt surgery and to identify clinical predictors of an improvement in terms of cognition and frailty during the same period. The main objective is to identify predictors of the effect on the planned shunt surgery on cognition and frailty.

Idiopathic normal pressure hydrocephalus (iNPH) is characterized by one or more of the symptoms gait disturbance, cognitive decline and urinary incontinence. The pathophysiology is not fully understood, but disturbed cerebrospinal fluid (CSF) circulation is considered one of the contributing factors. The CSF disturbance can be treated by shunt surgery with significant improvement in gait velocity and balance, whereas improvements in cognitive function are more uncertain.

Frailty is a consequence of cumulative decline in many physiological systems during a lifetime and implies an increased vulnerability to poor resolution of homoeostasis after a stressor event. Frailty is emerging as an important risk factor for mortality and postoperative complications but has to a limited degree been studied in iNPH.

We have previously, in a cross-sectional design, described cognitive profile and frailty status in patients with iNPH accepted for shunt surgery at Oslo University Hospital. The aims of this longitudinal follow-up of the same patient cohort are twofold:

1. to describe changes in cognitive profile and frailty from pre shunt to one year after the shunt surgery, and
2. to identify clinical predictors of an improvement in terms of cognition and frailty during the same period.

The patient sample consists of 276 patients that were accepted for shunt surgery at Oslo University Hospital in the period from September 2018 to December 2023. Their mean age was 73.1 years (range 52-85), 61% were men, and their mean length of education was 12.5 years. For the cognitive tests, we utilised z-scores (number of standard deviations (SD) from the age and education adjusted mean in a normative dataset).

For frailty, we used a 35 items Frailty Index (FI). Most of the items (frailty indicators) are scored 0 (not present) or 1 (present), while some of them have a graded score. The index is the sum score divided by the number of items, varying from 0.0 (no frailty) to 1.0 (extreme frailty). We used the same approach for assessment of the degree of frailty within each frailty domain. Cognitive decline is considered as one component of frailty. Accordingly, the Mini Mental State Examination (MMSE) is part of the FI and has a graded score. MMSE sum score <21 gives a score of 1 at this particular FI item, an MMSE score of 21-23 gives an item score of 0.7, an MMSE score of 24-26 gives an item score of 0.3, whereas an MMSE score of 27-30 does not contribute to the FI. MMSE is a screening test covering several cognitive domains. We will use MMSE to evaluate the cognitive dimension of frailty as described here, whereas more specific cognitive tests are utilised to assess failure in particular cognitive domains.

Preoperatively, the mean FI score was 0.23, and the most common frailty markers were in the domains of physical function and instrumental activities of daily living (iADL). On the group level, the iNPH patients were impaired on all cognitive domains preoperatively, but compared to patients with Alzheimer's disease, they had relatively better-preserved memory and more severe impairments in phonemic fluency.

220 (80%) patients were assessed one year postoperatively. The same cognitive tests and frailty assessments were carried out, making it possible to calculate simple change scores as score(postop) - score(preop). We will compare baseline characteristics of patients lost to follow-up descriptively with those followed, to assess potential attrition bias.

A detailed Statistical Analysis Plan (SAP) describes the planned analytical approach.

Inclusion Criteria:

• Diagnosed with iNPH and accepted for shunt surgery according to the American-European guidelines at Department of Neurosurgery, Oslo University Hospital, Rikshospitalet.

Exclusion Criteria:

• Non-native speakers of Norwegian
• Patients who had completed ≤ 3 cognitive tests

see inclusion criteria

The raw data supporting the conclsion of the article will be made available, following ethic committee guidelines on resonable request from any qualified investigators.

Data sets are available from the PI on request.