Cognitive profile of idiopathic normal pressure hydrocephalus (INPH)

Abstract

AbstractBackgroundCognitive deficits in people with idiopathic normal pressure hydrocephalus (INPH) are thought to differ from Alzheimer’s dementia (AD) but this has not been systematically explored.MethodWe undertook a systematic review and meta‐analysis. Peer‐reviewed studies publishing new data, with defined diagnostic criteria for INPH, and using formal cognitive tools were included. Quality was assessed using the Newcastle‐Ottawa Scale. Weighted mean difference (WMD) in score was calculated using random‐effects meta‐analysis for cognitive tests used in at least 2 studies. We assessed heterogeneity using I2. Sensitivity analyses explored results according to global cognition.Result13 studies met inclusion criteria. Participants with INPH had a pooled weighted MMSE score of 22.6 (95% confidence interval (CI) 21.3‐23.9). Studies were divided into those that matched AD participants to INPH participants for global cognition or selected mild AD (n = 6) and those that did not (n = 7). Participants with INPH performed better on the Boston Naming Task across all studies (n = 2; WMD: 1.41; 95% CI: 0.50, 2.31; I2 47%). There were no significant differences in semantic or phonemic fluency, delayed verbal recall, digit span backwards or forwards, or constructional praxis between AD and INPH across all studies. However, sensitivity analyses examining INPH participants compared to mild or matched AD showed significantly worse performance in INPH for semantic (n = 3; WMD: ‐1.37; 95% CI: ‐2.19, ‐0.55; I2 0%) and phonemic fluency (n = 3; WMD: ‐5.18; 95% CI: ‐6.95, ‐3.41; I2 0%), and in the Frontal Assessment Battery (n = 3; WMD: ‐2.33 points; 95% CI: ‐3.30, ‐1.36; I2 0%). Worse performance was also seen on sensitivity analysis in digit span backward (MD: ‐0.70; 95% CI: ‐1.20, ‐0.20) and constructional praxis (MD: ‐0.50; 95% CI: ‐0.90, ‐0.10) but only 1 study was available for each test.ConclusionIn this meta‐analysis, people with INPH tended to score within the mild dementia range on global cognitive testing. There were observable differences in cognitive profile between INPH and AD. People with INPH may have less aphasia but score significantly lower on tests of executive function than people with mild AD. These results could inform identification of INPH in memory services.