Subclinical Vascular Disease of the Brain in Dialysis Patients

Patients with chronic kidney failure treated by dialysis are at high risk for acute, clinically overt stroke. After accounting for age, sex, and race, there is a 3- to 9-times greater risk for hospitalization due to hemorrhagic and ischemic stroke among dialysis patients compared to the general population.1 In part, this excess burden of stroke may be related to accelerated cerebrovascular atherosclerosis2 or uncontrolled hypertension in uremia. Alternatively, factors related to the hemodialysis treatment itself, such as anticoagulation, rapid changes in blood pressure, or cerebral hypoperfusion induced by hemoconcentration,3 may increase the risk of stroke, as suggested by the high proportion of stroke events occurring during or soon after a dialysis treatment session.4

In addition to this risk of clinically apparent cerebrovascular disease, dialysis patients are also at greater risk for covert ischemic brain disease. One form of the latter is subclinical or “silent” stroke. These are previously unrecognized lesions on cranial imaging that appear as infarcts but are not associated with a clinical syndrome consistent with an acute stroke. These silent infarcts are common in older adults in the general population, and occur 5 times more frequently than clinically overt strokes.5 Despite the absence of acute neurological symptoms, the presence of such infarcts predicts a range of long-term adverse outcomes including incident dementia,6 clinically evident strokes,7, 8 and decline in physical and cognitive function. The Cardiovascular Health Study, a study of community-dwelling adults age 65 years and older, previously demonstrated that the prevalence of subclinical brain infarcts is inversely related to level of kidney function.9

Furthermore, dialysis patients are at especially high risk, with 5-times greater prevalence compared to those without kidney disease.10 Dialysis patients with subclinical infarcts also have a poor prognosis, with a 4-times higher risk of acute vascular events in models that adjust for baseline comorbid conditions.11

Another common radiographic finding that likely represents covert brain ischemia is leukoaraiosis, or white matter disease. Leukoaraiosis is defined radiographically as hyperintense signals in the white matter on T2-weighted and fluid-attenuated inverse recovery (FLAIR) magnetic resonance imaging, with normal-appearing signal on T1-weighted images. There is a growing consensus that white matter hyperintensities represent ultrastructural changes in brain white matter in response to chronic ischemia or hypoperfusion.12, 13 Hypertension is one of the major risk factors for leukoaraiosis, and strict blood pressure control may slow the progression of lesions.14 Other vascular risk factors such as older age, diabetes, increased thrombogenic factors, and extracranial atherosclerosis have also been variably linked to leukoaraiosis in the general population. The presence and severity of white matter hyperintensities have also been associated with a range of adverse outcomes. These include an increased risk of clinically evident stroke, even after adjustment for known stroke risk factors and for the presence of other brain imaging abnormalities.8, 15 Associations have also been reported with worse performance on tests of physical and motor function, including gait speed and fine motor function.16 Additionally, many authors have reported associations with poorer cognitive performance, especially on tests of executive function and motor speed.17, 18, 19, 20 These associations suggest that leukoaraiosis either directly affects brain function adversely or marks individuals at high vascular and cognitive risk.

Although prior studies have documented a high prevalence of leukoaraiosis in hemodialysis patients as well as patients during the earlier stages of chronic kidney disease,21, 22 Kim et al23 have added to the literature by documenting a 68.4% prevalence of leukoaraiosis in peritoneal dialysis patients from 3 units in Korea. This high prevalence is even more striking given that the patients were relatively young (mean age of 48 years), nondiabetic, and without clinical evidence of neurological disease. The data are strengthened by use of an age- and sex-matched nondialysis hypertensive control population whose prevalence of leukoaraiosis was only 17.5%. These data suggest that the presence of hypertension and factors related to hemodialysis treatment per se are not the only explanations of the high frequency of leukoaraisosis in dialysis patients; other factors related to uremia are also likely important. However, there are some limitations that need to be recognized in this study. First, both the peritoneal dialysis cohort as well as the hypertensive cohort are samples of convenience, rather than random samples; therefore, the generalizability of the results remains to be determined. Second, this is a relatively small cross-sectional study of prevalent peritoneal dialysis patients. As a result, one cannot reliably evaluate risk factors for incident leukoaraiosis, the role that survivor bias may play in the results, and the relationship of these findings to the prevalence of leukoaraiosis in incident dialysis patients. If indeed the prevalence of leukoaraiosis was also high in incident patients, this would support the hypothesis that factors related to dialysis treatment are not essential to cause the lesions.

In sum, the markedly high prevalence and severity of white matter disease documented by Kim et al in peritoneal dialysis patients provides important evidence for a high burden of silent ischemic brain disease in this population, with potentially important consequences for brain dysfunction. For example, it is now well recognized that there is a high prevalence of cognitive impairment in dialysis patients24; future studies should examine the contribution of subclinical ischemic vascular disease to this impairment by combining assessment of white matter disease burden with measures of cognitive and physical function. Future investigations in this area should consider other important methodological concerns. Recent advances in computer-guided image analysis allow for quantification of total and regional burden of affected white matter. Such semiautomated measurements of white matter disease have greater ability to discriminate those with neurological and psychological impairment, and are more sensitive to the longitudinal effects of interventions.25 Other novel magnetic resonance white matter imaging techniques, such as diffusion tensor imaging, may also be more sensitive at detecting early white matter injury in those at high vascular risk.26 Additionally, longitudinal studies are needed to determine whether white matter disease in dialysis patients is a static condition or progresses over time, and which factors—including those related to the dialysis treatment itself—predict more rapid progression. Such studies will be of crucial importance in identifying opportunities to preserve brain function in the rapidly growing population of dialysis patients.