Hypertension and white matter lesions of the brain

The incidence of cerebral white matter lesions (WML) is not uncommon in elderly people on magnetic resonance imaging (MRI). Age, history of hypertension and atherosclerotic cardiovascular disease have been identified as potential risk factors for WML, but results from previous studies have not been unequivocal. Thus, the aetiology of WML remains somewhat unclear. In this issue of the journal, Sierra and colleagues have investigated the influence of different components of high blood pressure on the presence of WML in an homogenous sample of asymptomatic, middle-aged patients who had never been treated for hypertension [1]. The brain white matter The normal white matter of the brain contains nerve fibres (axons), neuroglial cells (astrocytes and oligodendrocytes), vascular structures and interstitial space. The nerve structures in the white matter are mainly myelinated axons, which make up tracts leading to and from the brain, as well as connections between different areas in the brain. Myelin, a lipid-rich tissue containing large macromolecules, acts as an insulator around the axons, and increases the speed of conduction and improves its efficiency. The brain white matter is vulnerable to ischaemia. This can be explained by its particular arterial angioarchitecture as well as by the critical balance between blood flow and metabolic demands. An impaired autoregulation of the brain appearing, for example, with ageing, particularly in patients with hypertension, makes the brain (predominantly the white matter) more susceptible to episodes of hypoxia. Pathology and aetiology MRI has proved to be a very sensitive method for detecting any form of white matter abnormalities. Most pathological processes in the white matter will appear as hyperintensities in T2-weighted MR images, and thus MRI is unspecific and unable to distinguish between different types or grades of white matter damage. Accordingly, various neuropathological findings have been reported to correlate with histopathological findings in WML. The basic elements of white matter pathology are: (i) a loss of parenchymal elements (myelin, axons and glial cells), (ii) an increase of glial cells (gliosis), vascular changes, shrinkage/white matter atrophy, and (iii) necrosis with scarring or cavitation [2]. There is strong evidence supporting the hypothesis that atherosclerosis plays an important role in the pathogenesis of WML [2,3]. Extensive WML (i.e. patchy and confluent WML in the deep white matter) as well as irregular and extensive periventricular changes often have an ischaemic origin. Pathophysiologically, they can result from cerebral hypoperfusion and haemodynamic ischaemia secondary to atherosclerotic thickening of the penetrating arteries. Ageing and hypertension, the major risk factors for WML, are both related to atherosclerosis. It is suggested that the pathogenesis of WML comprises a variety of abnormalities from tiny punctuate or patchy lesions to extensive changes. Furthermore, the aetiology of changes in deep white matter and the periventricular area may not be the same due to the differences in their vascular supply [4]. Blood pressure and WML In the study by Sierra et al., WML were found to be common (41%) in asymptomatic and untreated middle-aged subjects with essential hypertension [1]. This study is important since it was one of the first times when 24 h ambulatory blood pressure monitoring (ABPM) was applied in a study of WML. People with WML had significantly higher values of systolic, diastolic, mean arterial pressure and pulse pressure not only based on the office blood pressure measurements, but also confirmed by 24 h ABPM. The severity of high blood pressure was found to be the most important factor related to the presence of WML. Thus, this study confirms the results from previous, less stringent studies where the presence of WML have been reported to be associated with the level of blood pressure and the lack of blood pressure control in hypertensive patients. The association between pulse pressure and WML is of interest. Pulse pressure has regained increasing interest lately [5] and has been suggested to be an independent risk factor for cerebrovascular and cardiovascular events and mortality [6–9]. However, much of the effect related to pulse pressure is related to systolic blood pressure [10]. Recently, cerebral WML have been found to be associated with decreased artery elasticity index [11]. Because elevated pulse pressure is usually regarded as a manifestation of increased arterial stiffness, this finding supports the true relationship between pulse pressure and WML. Whether elevated pulse pressure might be used as a marker of the risk of development of WML need to be clarified in further prospective studies. In the study by Sierra et al., no differences were observed in either nocturnal fall of blood pressure or in blood pressure variability, assessed by 24 h standard deviation, between patients with and without WML [1]. This is in contrast to earlier findings suggesting that abnormal diurnal blood pressure variation is correlated with the magnitude of WML [12,13]. This discrepancy may be related to the differences in study populations; Sierra et al. only included asymptomatic subjects without previous treatment for hypertension while, in the previous studies, long-standing hypertensive patients with vascular damage were included. Nevertheless, in the study by Sierra et al., a significantly lower nocturnal reduction in heart rate was found in patients with WML than in those without. This might be a reflection of an imbalance between sympathetic and parasympathetic nervous system activities, which may contribute to the development of cerebral lesions in hypertension. However, this novel finding needs to be specifically addressed in further studies. Methodological considerations A major problem concerning the study of WML and their aetiology arises from the variety of methodologies used [14]. The first, and probably the most important one, relates to the differences in the selection of study populations. Sierra et al. studied a homogenous group with a small age range (50–60 years) to better assess the true effect of blood pressure on the presence of WML. Thus, many confounding factors that could influence the outcome were excluded. The study subjects were never treated for hypertension, and thus the possible modifying effect of antihypertensive therapy could be controlled for. The fact that all the study subjects had hypertension, however, might have reduced the variability in some blood pressure parameters and explain why certain parameters of blood pressure (circadian pattern, long-term variability) did not predict the lesions. Another major problem in epidemiological studies of WML is the great number of rating scales used to grade these lesions. The agreement between the different rating scales is often poor, and when Mäntyläet al. compared 13 different scales, they concluded that some of the inconsistencies in results among the previous studies were actually due to differences in scales [15]. In addition, risk factors for WML (e.g. blood pressure measurement standardization, the definition of hypertension) have not been uniformly defined across studies, but different criteria were used. A graded association between a risk factor and the outcome (i.e. WML) is one of the most important pieces of proof for a causal link. Therefore, the standardization of risk factor measurements and their definition are necessary, together with comparable grading of WML. Most studies of risk factors of WML thus far have been cross-sectional. Their potential to investigate causality is limited, since it is not possible to assure that the risk factors actually preceded the occurrence of WML. Furthermore, risk factors identified in cross-sectional studies may be associated with WML themselves or with survival of subjects with WML. With regard to high blood pressure and the development of WML, the true causal relationship is supported by a few longitudinal studies, even though the baseline white matter status may not have been determined in these studies [16–18]. There are also plausible biological mechanisms supporting this association between high blood pressure and WML. Large prospective intervention studies are needed to investigate the aetiology of WML and the role of hypertension and blood pressure control more precisely. Clinical considerations The clinical significance of WML has been much debated. It has been demonstrated that WML are associated with impaired cognitive functions, in particular declined attention and speed of mental processing [19,20]. Interestingly, elevated blood pressure and other vascular risk factors have been suggested to play an important role in the development of Alzheimer's disease [21]. WML are often detected in patients with Alzheimer's disease [22] and may be related to accelerated cognitive decline in individuals with mild cognitive impairment [23]. WML may also have a prognostic significance for stroke and cardiac events, but the prospective data are still absent. The study of Sierra et al. revealed that WML already are a common feature in middle-aged untreated hypertensive patients [1]. Similarly, the prevalence of high blood pressure is known to be common in middle-age, and the prevalence of hypertension and WML increases with age. Thus, the finding that severity of blood pressure was the most important factor related to the presence of WML warrants that early and effective treatment of hypertension may have wide effects in preventing not only cerebrovascular, but also other neurological diseases and WML. This is also supported by a recent study where the risk of severe WML was significantly lower in subjects with a normal blood pressure and hypertensive patients treated with antihypertensive drugs compared to hypertensive patients treated but still having high blood pressure [18]. On the basis of these findings, it can be proposed that therapeutic interventions aimed at controlling hypertension might help prevent development and progression of WML. This is a promising start to identify improved strategies for neuroprotection in high-risk subjects, such as hypertensive patients. While active antihypertensive therapy is certainly important for every individual hypertensive patient, we need more information about the most efficient ways to prevent WML and the subsequent risks of cognitive deterioration.