Diagnosis, Treatment, and Analysis of Long-term Outcomes in Idiopathic Normal-Pressure Hydrocephalus

Abstract

To the Editor: We read with interest the article by McGirt et al. (9) entitled “Diagnosis, Treatment, and Analysis of Long-term Outcomes in Idiopathic Normal Pressure Hydrocephalus.” The authors conclude that gait impairment is the primary symptom that independently predicted improvement after shunting. The study included 132 patients, 129 (98%) of whom had gait impairment as a feature. We would like to comment on this particular selection criterion, as well as improvements in the Mini Mental State Examination (MMSE) as the sole measure of postoperative cognitive improvement. One of the inclusion criteria for this study was presentation with two or more features of the classic triad. Thus, the analysis inevitably includes patients presenting with dementia and urinary incontinence symptoms but not gait ataxia. Although dementia was originally thought to be the main feature of this syndrome, we have now come to understand that gait disturbance is an essential selection feature for any study with normal pressure hydrocephalus (NPH) patients (4, 6). In a prospective study of 151 patients, Marmarou et al. (8) recently advised that gait improvement immediately after external lumbar drainage is the best prognostic indicator of a positive shunt outcome with greater than 90% accuracy for prediction. Although the two extra inclusion criteria, namely either A- or B-waves, present during continuous pressure monitoring and clinical improvement in symptoms during a 3-day trial of controlled cerebrospinal fluid (CSF) drainage should safeguard against any false positive cases, we still think that the presence of gait ataxia should have been an essential inclusion criterion in the above study. Recent guidelines published in Neurosurgery and studies by other authors suggest that currently there is no evidence to support the prognostic value of A- or-B waves in patients with idiopathic NPH (7, 11, 14). Furthermore, in the same review, we saw that the positive predictive value of the external lumbar drainage averages 85% (7). In the case of McGirt et al. (9), only three patients did not have gait ataxia. Because the level of significance was set at less than 0.05, this small number does not produce a Type I error. The MMSE (5) is a general screening tool for cognitive impairment, dementia, and delirium consisting of five areas of assessment: orientation, registration, attention and calculation, recall, and language. The procedure takes little more than 10 minutes to conduct which, combined with its simplicity in terms of the minimal training required to administer the test, make it one of the most widely used research tools for the screening of cognitive impairment (13). However, there are numerous weaknesses associated with this method which become particularly pertinent when the MMSE is used as the sole measure of cognitive capacity, as was the case in McGirt et al.'s article. Problems associated with the use of MMSE are the lack of a published scoring manual, leading to differences in scoring and interpretation between users (1), variations in the use of cut-off scores as an indication of impairment, and the need to incorporate the educational level of the patients in analyses (12). The type of cognitive decline experienced in NPH varies widely; thus, it is essential that measures used to assess dementia-related aspects of the disorder are sensitive to the range of difficulties with which patients present. Such symptoms may include dyscalculia, acalculia (4), depression, and anxiety (10). Furthermore, the importance of matching all neuropsychological test scores with premorbid intelligence quotient levels and age-matched control data to obtain an accurate perception of any cognitive decline is highlighted by Devito et al. (2). The MMSE does not cover the above aspects of performance, and its ability to assess deficits in hydrocephalus is further reduced by the study's failure to match the data according to the criteria outlined. The MMSE was developed as a screening tool for cognitive decline and should only be used as such (3). As a screening tool, it has its merits; however, it should always be used in conjunction with, and not replace, a comprehensive neuropsychological assessment. Although its brevity makes it speedy and efficient as a screen, it consequently lacks depth and insight into the nature of the cognitive components that may be impaired. It is understood that McGirt et al. began recruiting patients in 1993 when our understanding of NPH was still evolving; thus, the results of the study are noteworthy, particularly as the authors report their results in a large sample size. However, with respect to the inclusion criteria, we emphasize gait as an essential selection criterion. Furthermore, we do not recommend the use of MMSE to provide an accurate account of the cognitive difficulties associated with NPH but rather suggest the use of a formal neuropsychology assessment. Andrew Tarnaris Rebecca F. Stephenson Lisa Cipolotti London, England