Review: neuropsychological test performance as an indicator of silent cerebrovascular disease in elderly hypertensives

Age and Ageing,  Nov, 1994  by L. Kalra,  S.H.D. Jackson,  C.G. Swift

• E-mail
• Print
• Link

Introduction

The physical consequences of untreated hypertension in elderly people, particularly stroke, heart failure and myocardial infarction, are well known [1-7]. Necropsy studies, however, suggest that the extent of central nervous system (CNS) involvement in hypertension may be underestimated during life. Significant pathological changes occur in the brain and cerebral blood vessels of at least 50% of hypertensive patients [8-10], although neurological symptoms and signs are identified in only 10-30% [11-13]. Recent magnetic resonance imaging (MRI) studies have shown significant white matter pathology in hypertensive patients without significant target organ damage (renal, retinal or cardiac involvement), especially in the older age group [14, 15]. The effects of hypertension on the CNS may be even greater than suggested by necropsy or MRI studies, because of alterations in CNS physiology and in the organization of brain activity during life which may be associated with little or no pathological changes.

The possibility of silent cerebrovascular disease being associated with hypertension was recognized more than a century ago [16, 17] and sub-clinical cerebral disorganization due to recurrent angiospastic episodes was predicted [18, 19]. Failure to recognize such CNS involvement in hypertension may have been largely due to the use of assessment procedures depending heavily on physical signs and limitations of investigations in detecting minimal changes in the brain substance. The effect of hypertension on CNS function has received little attention in the past. Interest in this area has been stimulated in recent years by the development of reliable and easily reproducible neuropsychological tests sensitive to minor changes in CNS function.

The objective of this paper is to review the growing literature on the cerebral effects of mild to moderate hypertension in the absence of clinically obvious cerebrovascular disease, with special reference to elderly patients. The potential physiological mechanisms underlying this association and the implications of these findings for the future management of hypertension in elderly people are explored.

Hypertension and Cerebral Function

Early studies: Early studies, using neuropsychological tests in hypertensive subjects, uncovered impairment in tasks of visual perception (flicker fusion threshold), motor speed (sustained finger tapping) and general cognitive functioning in these subjects [20-23]. These poorly controlled studies in small samples of patients of all ages (17-70 years) with essential or secondary hypertension were in no way conclusive but did suggest an association between hypertension and cerebral function which merited investigation.

Performance on neuropsychological batteries: Several studies undertaken in the following years used a range of neuropsychological tests and have shown impairment in several areas of cognitive functioning (Table I). Hypertensive subjects consistently showed impaired performance compared with normotensive subjects in timed tasks such as choice reaction time [24-28] (Table I), even when factors such as target organ damage and antihypertensive treatment were controlled for. Impaired performance has also been reported in various sub-tests of the Wechsler Adult Intelligence Scale (WAIS) battery in both newly detected and chronic hypertensive subjects [24, 27, 29-38]. Performance subscales (all timed tests) were more affected than verbal subscales (mostly untimed tests) suggesting impairment in attention, psychomotor speed and central information processing [32-36]. Hypertensive subjects also showed impaired performance on several sub-tests (e.g. Categories Test and the Trail Making Tests A and B) of the Halstead-Reitan battery when compared with normotensive controls [24, 29, 30, 33, 37-39]. These findings have been confirmed in unmedicated (n = 182) and in never-treated hypertensives in whom there was an inverse relationship between systolic and diastolic blood pressure and cognitive functioning as measured on a range of neuropsychological tests from the Halstead-Reitan battery [40]. [TABULAR DATA I OMITTED]

Performance on memory tests: In addition to impairment on tests requiring attention and psychomotor speed, hypertension has also been shown to affect anterograde memory tasks and thereby new learning [41-46]. Using a sensitive short-term memory search paradigm which allows discrimination between the different components of short-term memory (search speed, encoding, memory comparison and response selection), it was possible to show that this impairment was due to a slowing in the memory search speed in hypertensive subjects, rather than to changes in encoding and response processes [43, 44].

Limitations of neuropsychological studies: Despite several studies (Tables I and II) and reviews, the psychological aspects of hypertension [40, 46, 47], the extent and the nature of the association between mild to moderate essential hypertension and cognitive function remain equivocal. There are several reports (some on very large samples) which failed to show an association between hypertension and cognitive function [48-50]. These discrepancies may be due to differences in study designs which range from cross-sectional [24, 27, 45, 48, 50] or longitudinal [32-38, 41, 42] studies to group comparisons [25, 28, 31, 43]. The methodology of neuropsychological assessment has varied widely with tests being undertaken in non-standardized environments [41, 42, 48, 50] or under uniform conditions [28, 29, 31-38, 43! making comparisons between studies difficult. Clinical criteria for subject selection have differed with some studies including subjects with target organ changes [24, 25, 41, 42], or with blood pressure levels in the normotensive ranges [33-37, 49] or subjects on treatment in whom the CNS effects of drugs may have confounded changes due to hypertension [33-38, 41, 42]. Other problems include multiple versus single blood pressure measurements, use of averaged blood pressure levels versus single estimations and differences in time intervals between blood pressure measurement and neuropsychological assessment [51, 52]. The validity of analytic techniques used in processing data has also been questioned [51]. There is, however, emerging consensus that cognitive performance, especially on tasks involving attention and memory is impaired in hypertensive patients compared with normotensive subjects [42, 44, 46, 47, 51].