© 1997 American Heart Association, Inc.
Articles |
Cardiovascular Risk Factors and Hyalinization of Renal Arterioles at Autopsy
The Honolulu Heart Program
From the Honolulu Epidemiology Research Unit, Epidemiology and Biometry Program, Division of Epidemiology and Clinical Applications, National Heart, Lung, and Blood Institute (C.M.B.), and the Honolulu Heart Program, Kuakini Medical Center (P.-H.C.), Hawaii; and the Department of Pathology, Louisiana State University Medical Center, New Orleans (R.E.T., J.P.S.).
Correspondence to Cecil M. Burchfiel, PhD, Honolulu Heart Program, Thomas Sq Centre, 846 S Hotel St, Suite 306, Honolulu, HI 96813. E-mail buzz{at}hhp.hawaii-health.com
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Abstract Nephrosclerosis, commonly found in subjects with hypertension and diabetes, is marked by hyalinization of arterioles and fibroplasia of small arteries in the renal cortex. Cardiovascular risk factors that predicted subsequent hyalinization of renal arterioles at autopsy were identified, using data from the Honolulu Heart Program, a prospective epidemiological study of cardiovascular disease (CVD) in Japanese-American men. Among 8006 participants at baseline, 1381 died between 1965 and 1982; 285 of these had a protocol autopsy, and 150 had assessments of arteriolar hyalinization from renal tissue. Subjects were categorized into four groups on the basis of the number of hyalinized arterioles per square centimeter of renal tissue, and CVD risk factor levels and proportions were compared across these groups with the use of general linear models and logistic regression. Multivariate assessment using logistic regression demonstrated that diastolic blood pressure (DBP) and glucose level were positively associated and alcohol intake was negatively associated with an elevated degree of renal arteriolar hyalinization, independent of other CVD risk factors. The odds ratios for elevated hyalinization associated with a 10–mm Hg increase in DBP, a 20-mg/dL increase in glucose level, and a 30-mL/d increase in alcohol intake were 1.97 (95% confidence interval [CI]=1.24-3.12), 1.23 (95% CI=1.07-1.41), and 0.24 (95% CI=0.11-0.55), respectively. Associations were similar when prevalent cases of CVD were excluded and when autopsy selection bias was taken into account. Renal arteriolar hyalinization was also more strongly associated with atherosclerosis in the larger cerebral vessels (Spearman's r=.59, P<.001) than in the coronary arteries (r=.16, P=.073) and aorta (r =.24, P=.022). Hyalinization was significantly related to cardiovascular-renal mortality, and this association was accounted for by other CVD risk factors. These findings suggest that blood pressure, glucose level, and alcohol intake are independent predictors of hyalinization in renal arterioles and that this type of renal vasculopathy may be a marker for atherosclerosis in other vascular regions, particularly the cerebral vessels. The protective association involving alcohol and the possibility that renal arteriolar hyalinization may be an indicator of cerebral atherosclerosis may warrant investigation in other populations.
Key Words: Asian Americans • diabetes mellitus • atherosclerosis • hyaline substance • nephrosclerosis
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Both hypertension and diabetes mellitus have been recognized as important underlying causes of end-stage renal disease, the consequences of which are considerable and well documented.1 Subjects with these two disorders are also clearly at an increased risk of CVD,2 3 4 5 although reasons for this excess risk are only partially understood. Nephrosclerosis, one of the causes of end-stage renal disease and frequently found in subjects with hypertension and diabetes, is a condition characterized pathologically by hyalinization of arterioles and fibroplastic intimal thickening of small arteries in the kidney.6 7 8 These two types of renal vasculopathy are thought to represent distinctly separate processes occurring in different locations of the renal cortex, although some correlation between them appears to exist.7 While the etiology is uncertain,7 several CVD risk factors, such as hypertension,9 10 11 12 13 14 diabetes,11 13 15 16 17 and cigarette smoking,18 19 have been associated with these microvascular abnormalities.
In general, fewer studies have focused on hyalinization of arterioles compared with fibroplasia of small arteries. It appears that the association of blood pressure with small-artery fibroplasia is stronger than its association with arteriolar hyalinization in most6 8 16 but not all9 10 studies. In contrast, diabetes is associated with more severe hyalinization but no more arterial fibroplasia than would be expected after the concomitant elevation in blood pressure is taken into account.11 13 14 The role of other CVD risk factors, including obesity and body fat distribution, lipid levels, physical activity, dietary factors, and alcohol intake, in renal arteriolar hyalinization has not been well characterized.
Hyalinization of renal arterioles involves accumulation of an amorphous, eosinophilic, glassy substance within the vascular wall.15 Because endothelial cell damage appears to be involved in the initial stages of atherosclerosis,20 it is possible that hyaline masses within the intima might reflect early vascular injury. In addition, several recent studies demonstrated that arteriolar hyalinization in the renal cortex was positively associated with CHD mortality17 and with coronary atherosclerosis6 in adults aged 25 to 54 years, suggesting that hyalinization of renal arterioles may also be a marker for coronary artery disease.
Autopsy data from the Honolulu Heart Program provided the opportunity to examine associations of CVD risk factors with hyalinization of renal arterioles. Associations of fibroplasia of small renal arteries with blood pressure and atherosclerosis in coronary arteries and the aorta were examined previously in this cohort.21 The primary purpose of this investigation was to identify CVD risk factors assessed at baseline or during the follow-up period that were independent predictors of renal arteriolar hyalinization at autopsy. To further examine this measure of renal vasculopathy, its associations with well-established measures of atherosclerosis in the coronary arteries, the aorta, and cerebral vessels and with renal artery fibroplasia were also assessed.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The Honolulu Heart Program is a prospective epidemiological study of risk factors for CHD and stroke among 8006 middle-aged Japanese-American men who were initially examined between 1965 and 1968. Participants were residing on the island of Oahu, were born between 1900 and 1919, and were 45 to 68 years of age at the baseline examination. Three subsequent examinations for more than 80% of surviving members of the cohort were conducted an average of 2 (1968 to 1970), 6 (1971 to 1974), and 25 (1991 to 1993) years later. Between 1970 and 1982, a series of three additional examinations focusing on lipids and lipoproteins was also completed for a 30% random subsample of the participants. Details concerning recruitment and participation of study subjects have been published.22
Since 1965, comprehensive efforts have been made to ascertain incident CHD and stroke through both examinations and hospital surveillance. Deaths were identified through review of hospital records, obituaries, and death certificates. A committee reviewed available hospital summaries, death certificates, and autopsy reports and determined incident CVD events and causes of death according to standardized criteria published previously.23 24 Island hospitals, mortuaries, and medical examiners' offices were contacted daily to identify impending autopsies and obtain the heart, aorta, and cerebral vessels as well as other tissue specimens whenever possible.
Study Population
Among 8006 participants at baseline who were observed prospectively from 1965 through 1982, a total of 1381 deaths occurred. There were 285 autopsies performed in which kidney samples were available, and a standard protocol was used for assessment of coronary arteries and the aorta. Kidney tissue specimens embedded in paraffin were adequate in 155 of these autopsies. The study population for this investigation consisted of 150 subjects for whom measurements of hyalinization in renal arterioles were made.
Risk Factors
A comprehensive baseline examination included demographic, lifestyle, medication use, and medical history information, as well as anthropometric, physiological, and laboratory measurements. For some risk factors, such as blood pressure and diabetes, additional information was available from subsequent examinations. Details of the standardized procedures used in these examinations have been published.23 25 Briefly, the mean of several standard blood pressure measurements taken from the left arm in a sitting position after 5 minutes of rest was determined for each of the first three examinations (three measurements from the first two exams and two from the third), and an average of these means over all three examinations was computed. Both mean arterial blood pressure ([SBP+2(DBP)]/3) and pulse pressure (SBP-DBP) were also calculated. For most analyses, mean levels of SBP and DBP over the first three examinations were used. Any reported use of antihypertensive medications at these first three examinations was used for most analyses. BMI at baseline was calculated as weight in kilograms divided by height in meters squared. Subscapular and triceps skinfold thickness was measured with Lange calipers. Serum levels of cholesterol, triglyceride, and glucose were measured in a nonfasting state 1 hour after a 50-g glucose load. Dietary intake was estimated from a 24-hour recall. Alcohol intake (milliliters per day) was estimated on the basis of the usual frequency of consumption of beer, wine, sake, and hard liquor for past and current drinkers. Average amount, duration, and pack-years of smoking were determined for former and current smokers. An index of physical activity was based on a weighted average of the time spent in various levels of activity, similar to the index used in the Framingham study26 and the Puerto Rico Heart Health Program.27 Diabetes was considered present if either of the following criteria was met: (1) a self-reported history of or medical treatment for diabetes at any study examination or (2) a physician's diagnosis of or treatment for diabetes, identified by review of all hospital discharge summaries between 1965 and 1982.
Pathological Evaluation
All assessments of renal vasculopathy (hyalinization of arterioles and fibroplasia of small arteries) and atherosclerosis in the coronary arteries, aorta, and cerebral vessels were made using standardized protocols that involved blinding of the pathologist to the clinical history and risk factor profile for all subjects.
Renal tissue embedded in paraffin blocks was sent to Louisiana State University, where sections of 6-µm thickness were cut and stained with periodic acid–Schiff alcian blue–metanil yellow.28 The cortical tissue of each kidney section was scanned systematically by one of the authors (R.E. Tracy) under a 10x objective lens, and all hyalinized vessels were counted according to a previously published protocol.29 Hyaline masses within arteriolar vessel walls were clearly apparent as intense red-stained areas, a characteristic reaction that hyaline masses uniformly display with periodic acid–Schiff. Hyalinized arterioles and prearteriolar vessels <80 µm (outer diameter) were distinguished from other hyalinized structures by morphology. Most of the hyalinized vessels were afferent arterioles. Hyalinization of renal arterioles was assessed by counting the number of hyalinized vessels that were present within the calculated surface area of the tissue section and was expressed as counts of hyalinized arterioles per square centimeter of tissue.
Intimal fibroplasia of small arteries in the kidney was assessed as described in detail previously.21 Briefly, all arteries in a tissue section with an outer diameter of 80 to 300 µm were examined under a 10x objective lens. The outer diameter of the least axis of the elliptic profile was measured from one outer media to the other, excluding the adventitia. The thickness of the intima was also measured along the least axis by using a 40x lens. A ratio of intimal thickness to outer diameter was calculated for each artery, and the average of these ratios was used as an index of intimal fibroplasia. These ratios were determined separately for larger arteries located more proximal to the heart (outer diameter, 150 to 300 µm) and for smaller, more remote arteries (outer diameter, 80 to 149 µm).
The methods used to assess the degree of atherosclerosis in the coronary arteries, aorta, and cerebral vessels were described previously.30 31 32 33 By use of a standardized protocol, coronary arteries (left anterior descending, circumflex, and right) and the aorta were opened longitudinally, sewn to plastic sheets, and fixed in formalin. The extent of atherosclerosis in these vessels was determined by two methods, the American Heart Association panel method34 and an unaided visual estimation technique developed for the International Atherosclerosis Project.35 In the first method, a single pathologist compared the vessels with a panel of photographs, each indicating an increasing degree of atherosclerosis, and assigned a score ranging from 1 (completely free of raised lesions) to 7 (severe atherosclerosis with complicated lesions). In the second method, percentages of intimal artery surface area with fibrous plaques, complicated lesions, and calcified lesions were combined as a single category, representing the percentage of intimal surface area affected by raised lesions. For both measures of atherosclerosis, the mean of the three coronary artery estimates was used, since similar results were obtained for the arteries examined separately.
To estimate the degree of cerebral atherosclerosis, the circle of Willis and its major branches were attached to plastic sheets and stored in 10% formalin before being examined by a single pathologist.33 36 Sections of the vessels were examined at 22 predetermined sites. Scores were assigned at each of the 22 sites and ranged from 0 (no evidence of atherosclerosis) to 4 (a thick plaque resulting in >50% lumen narrowing or involving the entire vessel circumference). The mean score used in analysis was derived by dividing the total score by the number of sites examined.
Statistical Analysis
Because of the skewed nature of the distribution of renal hyalinization, variation in the median number of hyalinized arterioles per square centimeter was assessed across three categories of age at baseline and age at death, using the Kruskal-Wallis test based on Wilcoxon rank scores, available in the SAS NPAR1WAY procedure.37 Spearman's rank correlation coefficients were used to assess associations between CVD risk factors (continuous variables) and renal arteriolar hyalinization. The number of hyalinized renal arterioles per square centimeter was also categorized into four groups, with a similar number of subjects in each group. Age-adjusted mean levels of continuous variables and percentages of discrete variables were compared across categories of hyalinization using general linear models and logistic regression, respectively.37
Multiple logistic regression analysis was used to identify CVD risk factors that independently predicted an elevated degree of renal arteriolar hyalinization at autopsy. Since no precedent existed regarding a degree of hyalinization that was considered abnormal, three cut points were considered in defining this dichotomous outcome (>2.5 [n=81], >4.0 [n=67], and >5.5 [n=56] hyalinized arterioles per square centimeter). Since results were similar using all three definitions, results are presented here for the 4.0 cut point because assessment of model fit was slightly better and the number of subjects identified as having an elevated degree of hyalinization was not overly limited, particularly when analyses were further stratified or restricted. Results for the other two definitions are also described. Risk factors that were significantly associated with hyalinization when considered separately with adjustment for age were initially included in logistic regression models. As a measure of association, ORs and their 95% CIs were derived from these logistic models. Multiple linear regression analysis was also performed, using a square root transformation of the number of hyalinized arterioles per square centimeter6 16 17 (because of its skewed distribution) as the dependent variable, to confirm results obtained with logistic regression and to potentially enhance detection of associations with CVD risk factors.
Several additional logistic regression models were also considered. Interaction terms involving alcohol intake, glucose level, diabetes, DBP, and antihypertensive medication were not statistically significant, and therefore, results from the simpler models are reported here. Analyses were also conducted separately in nondiabetic and diabetic subjects. Because it was possible that prevalence of CHD or stroke at baseline might have led to alterations in CVD risk factor levels, analyses were repeated after excluding subjects with these diseases at baseline.
To ascertain whether an autopsy selection bias might exist, associations of CVD risk factors with hyalinization were reassessed among 88 of the 150 autopsied subjects who did not die from a cardiovascular or renal cause. Cardiovascular and renal causes of death (n=62) included CHD (n=40), cerebrovascular disease (n=9), aortic aneurysm (n=9), cardiomyopathy (n=1), and chronic renal disease (n=3). Analysis of this "basal" group has been useful in evaluating this type of bias in other autopsy studies.32 33 38
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The distribution of hyalinized arterioles per square centimeter was positively skewed, with mean=7.7, median=3.0, and SD=11.1. The degree of hyalinization did not vary significantly with age at baseline or age at death. Median values were slightly higher for subjects over age 60 (4.1/cm2) at baseline than for younger men (2.6/cm2 for 47 to 54 years and 2.4/cm2 for 55 to 60 years), and median values were slightly lower for subjects who were older at time of death (2.8/cm2 for 60 to 69 years and 2.7/cm2 for 70 to 81 years) than for men who died at a younger age (5.1/cm2 for 52 to 59 years).
Several CVD risk factors assessed at baseline were significantly correlated with hyalinization of renal arterioles at autopsy (Table 1
). Blood pressure (mean of three examinations over the first 6 years), measures of adiposity (BMI and subscapular skinfold thickness), and glucose level were all positively associated with hyalinization, while alcohol consumption was negatively associated with hyalinization. Several additional blood pressure indices provided similar results; correlation coefficients were .19 and .26 for baseline SBP and DBP values, respectively, .30 for mean arterial blood pressure, and .18 for pulse pressure. Other factors, including age, triglyceride level, percentage of calories from fat, amount and duration of cigarette smoking, and physical activity, were not significantly correlated with arteriolar hyalinization.
|
The distribution of hyalinized renal arterioles per square centimeter was divided into four categories, each with a similar number of subjects. Age-adjusted mean levels and percentages of selected CVD risk factors were compared across these four hyalinization categories (see the Figure). Statistically significant increases in age-adjusted mean levels of DBP (mean of three examinations), glucose, and BMI were observed with increasing degree of hyalinization (tests for trend were P=.0005, P=.0001, and P=.0004, respectively). Compared with the lowest hyalinization category, subjects in the highest category had mean levels of DBP, glucose, and BMI that were 9 mm Hg, 72 mg/dL, and 3 kg/m2 higher, respectively. The percentage of subjects who reported taking antihypertensive medication at any of the first three examinations and who had diabetes during follow-up increased significantly with increasing hyalinization (P=.038 and P=.0001, respectively). An inverse trend was observed for mean alcohol intake among past and current drinkers and the degree of intimal hyalinization (P=.0023), with a 25 mL/d difference observed for lowest versus highest hyalinization categories. Statistically significant trends were also observed for subscapular skinfold thickness, total cholesterol level, SBP and DBP at baseline, SBP averaged over three examinations, and mean arterial blood pressure (data not shown). Factors that did not vary significantly across hyalinization categories included physical activity, triglyceride level, hematocrit, percent of calories from fat, fish consumption, smoking status, and average amount and duration of cigarette smoking (data not shown).
|
To identify CVD risk factors that were independently associated with hyalinization of renal arterioles, multiple logistic regression analyses were performed, using several definitions to represent elevated degrees of hyalinization (>2.5, >4.0, and >5.5 hyalinized renal arterioles per square centimeter). ORs and 95% CIs for risk factors associated with elevated renal arteriolar hyalinization defined with a cut point of >4.0 are provided in Table 2. Mean DBP and glucose level were positively associated and alcohol was negatively associated with an elevated degree of renal arteriolar hyalinization after adjusting for other variables in the model. Increases of 10 mm Hg in mean DBP and 20 mg/dL in glucose level were associated with 97% and 23% increases, respectively, in the odds of having elevated hyalinization. Similarly, a 30-mL/d increase in alcohol consumption was associated with a 76% decrease in the likelihood of having elevated hyalinization. Age at death was associated at borderline significant levels, while diabetes, antihypertensive medication, and BMI were not independently associated with hyalinization.
|
Several related analyses were also performed. Results were similar when cut points of >2.5 and >5.5 hyalinized arterioles per square centimeter were used as dependent variables in logistic regression (data not shown). Logistic models that included total cholesterol and triglyceride level were also considered, but neither variable was significantly associated with hyalinization, and therefore, these variables were excluded in favor of a more parsimonious model. When analyses were restricted to 87 nondiabetic subjects, DBP and alcohol intake were still independently associated with hyalinization, and the magnitude of the OR for glucose level was unchanged but no longer statistically significant (data not shown). Among the remaining 61 diabetic subjects, glucose level was a significant predictor of hyalinization, but associations involving DBP and alcohol were small in magnitude and nonsignificant. Multiple linear regression analysis was also conducted using a square root transformation of the number of hyalinized arterioles per square centimeter as the dependent variable to retain the continuous nature of this outcome variable. Significant independent predictors of renal hyalinization were unchanged and included mean DBP, glucose level, and alcohol intake.
Logistic regression analyses were repeated for a subset of the population that was free of prevalent CHD or stroke at baseline (Table 3). Results were essentially unchanged, with glucose level, mean DBP, and alcohol as significant independent predictors of hyalinization. When the analysis was restricted to subjects with a non-CVD and nonrenal underlying cause of death to assess potential autopsy selection bias, the magnitude of these associations remained similar, although the sample size was reduced and the probability values were correspondingly larger. For example, the ORs for men without exclusions (Table 2
) compared with those for men excluding CVD and renal mortality (Table 3) were 1.23 and 1.28 for glucose, 1.97 and 1.87 for DBP, and 0.24 and 0.32 for alcohol, respectively.
|
Additional analyses were conducted to assess whether the measure of renal arteriolar hyalinization used in this study was related to established measures of atherosclerosis in the heart, aorta, and brain and to an indicator of renal artery intimal fibroplasia (Table 4). While measures of atherosclerosis (AHA panel score and percent of intimal surface with raised lesions) were correlated with renal arteriolar hyalinization, stronger associations were observed for cerebrovascular atherosclerosis at the circle of Willis and its major branches (r=.59) and for the ratios of intimal thickness to outer diameter involving renal arteries relatively proximal (150 to 300 µm) to (r=.42) and remote (80 to 149 µm) from (r=.52) the heart. These measures showed progressive increases in age-adjusted mean levels with increasing degree of hyalinization (P<.0001). In addition, an elevated degree of renal arteriolar hyalinization was associated with a significantly elevated risk of mortality due to cardiovascular and renal causes (OR=2.64, 95% CI=1.34-5.20). This association was attenuated to nonsignificant levels with adjustment for major CVD risk factors (age, DBP, antihypertensive medication, cholesterol level, and pack-years of smoking; OR=1.91, 95% CI=0.91-4.02) and further attenuated with additional control for BMI, alcohol, and glucose level (OR=1.33, 95% CI=0.56-3.15).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Results from this autopsy study indicate that several CVD risk factors are independently associated with an elevated degree of hyalinization in renal arterioles. Glucose level and DBP were positively and significantly associated with hyalinization, while alcohol consumption showed a significant protective association. In general, the associations involving indicators of glucose intolerance and hypertension tend to confirm previous findings, while the independent associations involving alcohol intake have not been previously reported to our knowledge in other population-based autopsy investigations.
The independent association of blood pressure with arteriolar hyalinization is consistent with several other studies.9 10 14 16 39 Associations between elevated glucose or diabetes and arteriolar hyalinization are well recognized.11 13 14 15 16 17 The associations involving glucose level and DBP in the current study persisted when treatment for hypertension and diabetes, respectively, were also taken into account.
The inverse association of alcohol intake with renal arteriolar hyalinization is consistent with the protective associations of moderate alcohol consumption reported with clinical manifestations of CHD.40 41 42 The protective association of alcohol with CHD is thought to be due to beneficial effects of alcohol on high-density lipoprotein cholesterol levels.41 43 Associations of alcohol intake with atherosclerosis assessed at autopsy are inconsistent.44 For example, alcohol intake was inversely associated with raised coronary lesions in the International Atherosclerosis Project,45 but no significant association was found in a previous autopsy study of men from the current cohort.31 The association between alcohol and arteriolar hyalinization in the present study showed a dose-dependent relation that was independent of other CVD risk factors. High-density lipoprotein cholesterol measurements were not available at the baseline examination of this study.
BMI and subscapular skinfold thickness were associated with hyalinization of renal arterioles in univariate analyses, but multivariate assessment suggested that their association was accounted for by other CVD risk factors. It is possible that these measures were associated (noncausally) with hyalinization because of their relation with blood pressure and diabetes. Another possibility is that obesity may adversely affect blood pressure and glucose intolerance, which could then contribute to increased hyalinization. If this were true, including blood pressure and glucose level (intervening variables) in a logistic model would be expected to underestimate an association between obesity andrenal hyalinization.
Several other CVD risk factors were also examined in relation to hyalinization of renal arterioles. Age was not significantly related to hyalinization, in contrast to some studies14 17 46 but not all.16 A few previous studies have reported associations of cigarette smoking with microvascular abnormalities in several organs,47 including the kidney,18 19 although a dose-response effect was not demonstrated in one of these studies.18 Other investigations, including the present one, did not find a significant association with smoking.17 Serum cholesterol in this study was associated with renal hyalinization, consistent with results from one other study17 ; however, in multivariate analyses this association was accounted for by other CVD risk factors in the model.
Information on CVD risk factors was collected before assessment of hyalinization of renal arterioles at autopsy. An advantage of this design is the lack of bias that could otherwise result if risk factor data were collected retrospectively. A potential disadvantage is the possibility that risk factor values could change over time and thus result in some degree of misclassification. In this study we have combined information on blood pressure level and treatment over the first three examinations. In addition to reducing misclassification of blood pressure treatment status, the averaging of blood pressure over these three examinations may have reduced measurement error (particularly for SBP). Because of the relatively high incidence of diabetes in this cohort,48 information from examinations and hospital surveillance during the follow-up period was used to enhance classification of diabetic status. It is likely that any remaining misclassification of CVD risk factors would have been nondifferential and therefore would have diminished any of the associations observed.
Because autopsies are not performed randomly in a population, it is possible that certain associations will be observed simply because individuals who have both the risk factor of interest and the outcome being studied could be more likely to have received an autopsy. This potential bias has been extensively studied in this cohort, and evidence from investigations of CVD risk factors and atherosclerosis suggests that this type of selection bias has been minimal or nonexistent.31 32 33 In the present study, the strength of associations between CVD risk factors and renal hyalinization remained similar (although level of significance was reduced due perhaps to a reduction in sample size) when the autopsy sample was further restricted to the basal group,32 38 composed of subjects who did not die from cardiovascular or renal causes. This approach is also analogous to that described for addressing clinical selection bias in necropsies.49 These results suggest that the associations demonstrated in this study are not likely to be due to autopsy selection bias.
Endothelial cell damage may be involved in the initial stages of atherosclerosis.20 This early vascular injury may be associated with changes in permeability of the endothelial wall.15 50 51 52 Increased permeability of arterioles to macromolecules has been demonstrated in subjects with hypertension and diabetes.53 In addition, increased permeability of vessel walls in subjects with diabetes has been proposed as a mechanism that could account for their excess CVD risk.54 It is possible that these changes in permeability or necrosis of smooth muscle cells, as others have proposed,55 may be associated with deposition of hyaline within the vascular wall.
Hyaline, an amorphous, eosinophilic, glassy substance, is found in the small arteries and arterioles of a number of organs, most commonly in the spleen and kidneys, rarely in the heart and lungs, but also frequently in the brain, retina, and liver.15 The hyaline substance is composed of several constituents, including the third component of complement (iC3b in particular),15 hyaluronic acid,15 cholesterol,56 and fibrin in smaller amounts.56 Although hyalinization of arterioles is often referred to as "arteriolosclerosis," the hyaline masses tend to be easily deformed by pressure perfusion,57 consistent with the concept that these masses may result in a softening rather than a hardening of the arteriolar wall.7 16
Renal arteriolar hyalinization was also associated with measures of atherosclerosis in coronary arteries, the aorta, and major cerebral vessels; fibroplasia in renal arteries; and cardiovascular-renal mortality. It is possible that hyalinization in renal arterioles may be a marker for atherosclerosis in other vascular beds, analogous to findings from the Pathobiological Determinants of Atherosclerosis in Youth (PDAY) study, which showed that renal artery intimal thickness or fibroplasia was strongly associated with atherosclerotic raised lesions in coronary arteries and the aorta.58 Consistent with this hypothesis, hyalinization of renal arterioles was associated with both CHD mortality17 and coronary atherosclerosis6 among adults 25 to 54 years of age from the Community Pathology Study. Associations between hyalinization and raised lesions in the coronary arteries and aorta were slightly weaker and less consistent in the present study than in the Community Pathology Study that involved somewhat younger adults.6
The extent of renal arteriolar hyalinization was strongly correlated with a cerebrovascular atherosclerosis score (r=.59), and these scores showed an increasing trend with increasing degree of hyalinization. Several other studies have suggested that microvascular changes in the kidney may parallel those found in the brain.14 50 59 An autopsy study demonstrated that the medial hypertrophy associated with chronic systemic hypertension in small renal arteries corresponded with that observed in small cerebral vessels.59 Findings from the present study are also consistent with results from the Hisayama study, a prospective epidemiological study in Japan characterized by a high autopsy rate.14 Renal arteriolar hyalinization was significantly associated with mortality from cerebrovascular disease but not from CHD.14 It is possible that a common antecedent factor such as hypertension50 or glucose intolerance might lead to both arteriolar hyalinization and atherosclerosis in multiple locations. Renal hyalinization was also associated with cardiovascular-renal mortality in the current study. Since this association was not independent of other CVD risk factors, it appears that arteriolar hyalinization may reflect an adverse CVD risk factor profile.
The findings from this study suggest that blood pressure, glucose level, and alcohol intake may be important independent determinants of the extent of hyalinization in renal arterioles. Increases in blood pressure and glucose level were associated with significant increases in the likelihood of having an elevated degree of arteriolar hyalinization, while alcohol intake showed a strong protective association. While the consequences of increased hyalinization of renal arterioles have not yet been fully elucidated, results from this investigation indicate a relatively strong association with the extent of cerebrovascular atherosclerosis and with cardiovascular-renal mortality in general. These findings suggest that hyalinization of renal arterioles could be a marker for cerebral atherosclerosis in particular. Further investigation of the potential protective role of alcohol and hyalinization of renal arterioles as an indicator of atherosclerosis in other vascular regions may be warranted.
|
Selected Abbreviations and Acronyms |
|---|
|
|
Acknowledgments |
|---|
This study was supported by contracts NO1-HC-02901 and NO1-HC-05102 from the National Heart, Lung, and Blood Institute, Bethesda, Md.
Received May 1, 1996; accepted June 20, 1996.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. US Renal Data System. USRDS 1994 Annual Data Report. Bethesda, Md: National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases; 1994. NIH publication No. 94-3176.
2. Whelton PK. Epidemiology of hypertension. Lancet. 1994;344:101-106.[Medline] [Order article via Infotrieve]
3. Glynn RJ, Field TS, Rosner B, Hebert PR, Taylor JO, Hennekens CH. Evidence for a positive linear relation between blood pressure and mortality in elderly people. Lancet. 1995;345:825-829.[Medline] [Order article via Infotrieve]
4. Fuller JH, Shipley MJ, Rose G, Jarrett RJ, Keen H. Mortality from coronary heart disease and stroke in relation to degree of glycemia: the Whitehall study. Br Med J. 1983;287:867-870.
5.
Kannel WB, McGee DL. Diabetes and cardiovascular disease: the Framingham study. JAMA. 1979;241:2035-2038.
6. Tracy RE, Strong JP, Newman WP III, Malcom GT, Oalmann MC, Guzman MA. Renovasculopathies of nephrosclerosis in relation to atherosclerosis at ages 25 to 54 years. Kidney Int. 1996;49:564-570.[Medline] [Order article via Infotrieve]
7. Tracy RE. Nephrosclerosis from childhood to old age: a viewpoint. Geriatr Nephrol Urol. 1992;1:201-211.
8. Tracy RE, Bhandaru SY, Oalmann MC, Guzman MA, Newman WP III. Blood pressure and nephrosclerosis in black and white men and women aged 25 to 54. Mod Pathol. 1991;4:602-609.[Medline] [Order article via Infotrieve]
9. Yamaguchi T, Omae T, Katsuki S. Quantitative determination of renal vascular changes related to age and hypertension. Jpn Heart J. 1969;10:248-258.[Medline] [Order article via Infotrieve]
10. Katafuchi R, Takebayashi S. Morphometrical and functional correlations in benign nephrosclerosis. Clin Nephrol. 1987;28:238-243.[Medline] [Order article via Infotrieve]
11. Smith JP. Hyaline arteriolosclerosis of the kidney. J Pathol Bacteriol. 1955;69:147-168.[Medline] [Order article via Infotrieve]
12. Takazakura E, Sawabu N, Handa A, Takada A, Shinoda A, Takeuchi J. Intrarenal vascular changes with age and disease. Kidney Int. 1972;2:224-230.[Medline] [Order article via Infotrieve]
13. Tracy RE, Toca VT. Nephrosclerosis and blood pressure, II: reversibility of proliferative arteriosclerosis. Lab Invest. 1974;30:30-34.[Medline] [Order article via Infotrieve]
14. Ueda K, Omae T, Hirota Y, Takeshita M, Hiyoshi Y, Nakamura Y, Katsuki S. Epidemiological and clinico-pathological study on renal diseases observed in the autopsy cases in Hisayama population, Kyushu Island, Japan. J Chron Dis. 1976;29:159-173.[Medline] [Order article via Infotrieve]
15. Gamble CN. The pathogenesis of hyaline arteriolosclerosis. Am J Pathol. 1986;122:410-420.[Abstract]
16. Tracy RE. Blood pressure related separately to parenchymal fibrosis and vasculopathy of the kidney. Am J Kidney Dis. 1992;20:124-131.[Medline] [Order article via Infotrieve]
17. Tracy RE, Malcom GT, Oalmann MC, Newman WP III, Guzman MA. Nephrosclerosis, glycohemoglobin, cholesterol, and smoking in subjects dying of coronary heart disease. Mod Pathol. 1994;7:301-309.[Medline] [Order article via Infotrieve]
18. Black HR, Zeevi GR, Silten RM, Walker Smith GJ. Effect of heavy cigarette smoking on renal and myocardial arterioles. Nephron. 1983;34:173-179.[Medline] [Order article via Infotrieve]
19. Oberai B, Adams CWM, High OB. Myocardial and renal arteriolar thickening in cigarette smokers. Atherosclerosis. 1984;52:185-190.[Medline] [Order article via Infotrieve]
20. Fuster V, Badimon L, Badimon JJ, Chesebro JH. The pathogenesis of coronary artery disease and the acute coronary syndromes. N Engl J Med. 1992;326:242-250.[Medline] [Order article via Infotrieve]
21. Tracy RE, MacLean CJ, Reed DM, Hayashi T, Gandia M, Strong JP. Blood pressure, nephrosclerosis, and age: autopsy findings from the Honolulu Heart Program. Mod Pathol. 1988;1:420-427.[Medline] [Order article via Infotrieve]
22. Worth RM, Kagan A. Ascertainment of men of Japanese ancestry in Hawaii through World War II Selective Service registration. J Chron Dis. 1970;23:389-397.[Medline] [Order article via Infotrieve]
23.
Yano K, Reed DM, McGee DL. Ten-year incidence of coronary heart disease in the Honolulu Heart Program: relationship to biological and lifestyle characteristics. Am J Epidemiol. 1984;119:653-666.
24.
Kagan A, Popper JS, Rhoads GG. Factors related to stroke incidence in Hawaii Japanese men: the Honolulu Heart Study. Stroke. 1980;11:14-21.
25. Kagan A, Harris BR, Winkelstein W Jr, Johnson KG, Kato H, Syme SL, Rhoads GG, Gay ML, Nichaman MZ, Hamilton HB, Tillotson J. Epidemiologic studies of coronary heart disease and stroke in Japanese men living in Japan, Hawaii and California: demographic, physical, dietary and biochemical characteristics. J Chron Dis. 1974;27:345-364.[Medline] [Order article via Infotrieve]
26.
Kannel WG, Sorlie PD. Some health benefits of physical activity: the Framingham study. Arch Intern Med. 1979;139:857-861.
27. Garcia-Palmieri MR, Costas R, Cruz-Vidal M, Sorlie PD, Havlik RJ. Increased physical activity: a protective factor against heart attack in Puerto Rico. Am J Cardiol. 1982;50:759-765.
28. Tracy RE. Correlation of lengthy hospital records of blood pressure with nephrosclerosis. Am J Epidemiol. 1970;91;32-37.
29. Tracy RE, Malcom GT, Oalmann MC, Qureshi U, Ishii T, Velez-Duran M. Renal microvascular features of hypertension in Japan, Guatemala, and the United States. Arch Pathol Lab Med. 1992;116:50-55.[Medline] [Order article via Infotrieve]
30. Stemmermann GN, Steer A, Rhoads GG, Lee K, Hayashi T, Nakashima T, Keehn R. A comparative pathology study of myocardial lesions and atherosclerosis in Japanese men living in Hiroshima, Japan and Honolulu, Hawaii. Lab Invest. 1976;34:592-600.[Medline] [Order article via Infotrieve]
31.
Reed DM, MacLean CJ, Hayashi T. Predictors of atherosclerosis in the Honolulu Heart Program, I: biologic, dietary, and lifestyle characteristics. Am J Epidemiol. 1987;126:214-225.
32.
Burchfiel CM, Reed DM, Marcus EB, Strong JP, Hayashi T. Association of diabetes mellitus with coronary atherosclerosis and myocardial lesions: an autopsy study from the Honolulu Heart Program. Am J Epidemiol. 1993;137:1328-1340.
33.
Reed DM, Resch JA, Hayashi T, MacLean C, Yano K. A prospective study of cerebral artery atherosclerosis. Stroke. 1988;19:820-825.
34.
McGill HC Jr, Brown BW, Gore I, McMillan GC, Patterson JC, Pollack OJ, Roberts JC Jr, Wissler RW. Grading human atherosclerosis lesions using a panel of photographs. Circulation. 1968;37:455-459.
35. Guzman MA, McMahan CA, McGill HC, Strong JP, Tejada C, Restrepo C, Eggen DA, Robertson WB, Solberg LA. Selected methodologic aspects of the International Atherosclerosis Project. Lab Invest. 1968;18:479-497.[Medline] [Order article via Infotrieve]
36.
Mitsuyama Y, Thompson LR, Hayashi T, Lee KK, Keehn RJ, Resch JA, Steer A. Autopsy study of cerebrovascular disease in Japanese men who lived in Hiroshima, Japan, and Honolulu, Hawaii. Stroke. 1979;10:389-395.
37. SAS Institute Inc. SAS/STAT User's Guide, Version 6. Cary, NC: SAS Institute Inc; 1990.
38. Robertson WB, Strong JP. Atherosclerosis in persons with hypertension and diabetes mellitus. Lab Invest. 1968;18:538-551.[Medline] [Order article via Infotrieve]
39. Freedman BI, Iskandar SS, Appel RG. The link between hypertension and nephrosclerosis. Am J Kidney Dis. 1995;25:207-221.[Medline] [Order article via Infotrieve]
40. Rimm EB, Giovannucci EL, Willett WC, Colditz GA, Ascherio A, Rosner B, Stampfer MJ. Prospective study of alcohol consumption and risk of coronary disease in men. Lancet. 1991;338:464-468.[Medline] [Order article via Infotrieve]
41.
Gaziano JM, Buring JE, Breslow JL, Goldhaber SZ, Rosner B, VanDenburgh M, Willett W, Hennekens CH. Moderate alcohol intake, increased levels of high-density lipoprotein and its subfractions, and decreased risk of myocardial infarction. N Engl J Med. 1993;329:1829-1834.
42.
Goldberg RJ, Burchfiel CM, Reed DM, Wergowske G, Chiu D. A prospective study of the health effects of alcohol consumption in middle-aged and elderly men: the Honolulu Heart Program. Circulation. 1994;89:651-659.
43. Suh IL, Shaten BJ, Cutler JA, Kuller LH. Alcohol use and mortality from coronary heart disease: the role of high-density lipoprotein cholesterol. Ann Intern Med. 1992;116:881-887.
44. Moore RD, Pearson TA. Moderate alcohol consumption and coronary artery disease. Medicine (Baltimore). 1986;65:242-267.[Medline] [Order article via Infotrieve]
45. Moore MC, Guzman MA, Schilling PE, Strong JP. Dietary-atherosclerosis study on deceased persons. J Am Diet Assoc. 1975;67:22-28.[Medline] [Order article via Infotrieve]
46. Tracy RE, Heigle TJ, Velez-Duran M. Evidence for the failure of the Laplace law as a sole explanation for wall thickening of arteries in hypertensive and aging normotensive kidneys. Arch Pathol Lab Med. 1989;113:342-349.[Medline] [Order article via Infotrieve]
47. Auerbach O, Hammond EC, Garfinkel L. Thickening of walls of arterioles and small arteries in relation to age and smoking habits. N Engl J Med. 1968;278:980-984.
48. Burchfiel CM, Curb JD, Rodriguez BL, Yano K, Hwang L-J, Fong K-O, Marcus EB. Incidence and predictors of diabetes in Japanese-American men: the Honolulu Heart Program. Ann Epidemiol. 1995;5:33-43.[Medline] [Order article via Infotrieve]
49.
McFarlane MJ, Feinstein AR, Wells CK, Chan CK. The `epidemiologic necropsy': unexpected detections, demographic selections, and changing rates of lung cancer. JAMA. 1987;258:331-338.
50. Garcia JH, Ho K-L. Pathology of hypertensive arteriopathy. Neurosurg Clin N Am. 1992;3:497-507.
51.
Stary HC, Blankenhorn DH, Chandler AB, Glagov S, Insull W Jr, Richardson M, Rosenfeld ME, Schaffer SA, Schwartz CJ, Wagner WD, Wissler RW. A definition of the intima of human arteries and of its atherosclerosis-prone regions: a report from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. Circulation. 1992;85:391-405.
52. Nag S, Robertson DM, Dinsdale HB. Quantitative estimate of pinocytosis in experimental acute hypertension. Acta Neuropathol (Berl). 1979;46:107-116.[Medline] [Order article via Infotrieve]
53. Parving HH. Microvascular permeability to plasma proteins in hypertension and diabetes in man: on the pathogenesis of hypertensive and diabetic microangiopathy. Dan Med Bull. 1975;22:217-233.[Medline] [Order article via Infotrieve]
54. Keen H, Ashton CE. Mechanisms of excess cardiovascular mortality in diabetes. Postgrad Med J. 1989;65(suppl 1):S26-S29.
55. Montgomery PO'B, Muirhead EE. A characterization of hyaline arteriolar sclerosis by histochemical procedures. Am J Pathol. 1954;30:521-531.
56. Adams CWM, Bayliss OB, Orton CC. Plasma protein accumulation in arterial degenerations. J Atheroscler Res. 1967;7:473-489.[Medline] [Order article via Infotrieve]
57. Tracy RE, Overll EO. Arterioles of perfusion-fixed hypertensive and aged kidneys. Arch Pathol. 1966;82:526-534.[Medline] [Order article via Infotrieve]
58.
McGill HC Jr, Strong JP, Tracy RE, McMahan CA, Oalmann MC, and the Pathobiological Determinants of Atherosclerosis in Youth (PDAY) Research Group. Relation of a postmortem renal index of hypertension to atherosclerosis in youth. Arterioscler Thromb Vasc Biol. 1995;15:2222-2228.
59. Cook TA, Yates PO. A histometric study of cerebral and renal arteries in normotensives and chronic hypertensives. J Pathol. 1972;108:129-135.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  S. L. White, K. R. Polkinghorne, A. Cass, J. E. Shaw, R. C. Atkins, and S. J. Chadban Alcohol consumption and 5-year onset of chronic kidney disease: the AusDiab study Nephrol. Dial. Transplant., August 1, 2009; 24(8): 2464 - 2472. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ninomiya, M. Kubo, Y. Doi, K. Yonemoto, Y. Tanizaki, K. Tsuruya, K. Sueishi, M. Tsuneyoshi, M. Iida, and Y. Kiyohara Prehypertension Increases the Risk for Renal Arteriosclerosis in Autopsies: The Hisayama Study J. Am. Soc. Nephrol., July 1, 2007; 18(7): 2135 - 2142. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. S. Schaeffner, T. Kurth, P. E. de Jong, R. J. Glynn, J. E. Buring, and J. M. Gaziano Alcohol Consumption and the Risk of Renal Dysfunction in Apparently Healthy Men Arch Intern Med, May 9, 2005; 165(9): 1048 - 1053. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Y. Wong, R. Klein, A. R. Sharrett, B. B. Duncan, D. J. Couper, B. E.K. Klein, L. D. Hubbard, F. J. Nieto, and for the Atherosclerosis Risk in Communities Study Retinal Arteriolar Diameter and Risk for Hypertension Ann Intern Med, February 17, 2004; 140(4): 248 - 255. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. L. Knight, M. J. Stampfer, E. B. Rimm, S. E. Hankinson, and G. C. Curhan Moderate alcohol intake and renal function decline in women: a prospective study Nephrol. Dial. Transplant., August 1, 2003; 18(8): 1549 - 1554. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. L. Knight, M. J. Stampfer, E. B. Rimm, S. E. Hankinson, and G. C. Curhan Moderate alcohol intake and renal function decline in women: a prospective study Nephrol. Dial. Transplant., August 1, 2003; 18(88): 1549 - 1554. [Abstract] [Full Text]
|
|
|
|
|
|
C. Zoccali, F. Mallamaci, and P. Finocchiaro Atherosclerotic Renal Artery Stenosis: Epidemiology, Cardiovascular Outcomes, and Clinical Prediction Rules J. Am. Soc. Nephrol., November 1, 2002; 13(90003): S179 - 183. [Abstract] [Full Text]
|
|
|
|
 |
|
 T. Y. Wong, R. Klein, A. R. Sharrett, M. I. Schmidt, J. S. Pankow, D. J. Couper, B. E. K. Klein, L. D. Hubbard, B. B. Duncan, and for the ARIC Investigators Retinal Arteriolar Narrowing and Risk of Diabetes Mellitus in Middle-aged Persons JAMA, May 15, 2002; 287(19): 2528 - 2533. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Y. Wong, R. Klein, A. R. Sharrett, B. B. Duncan, D. J. Couper, J. M. Tielsch, B. E. K. Klein, and L. D. Hubbard Retinal Arteriolar Narrowing and Risk of Coronary Heart Disease in Men and Women: The Atherosclerosis Risk in Communities Study JAMA, March 6, 2002; 287(9): 1153 - 1159. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Ribstein, G. Du Cailar, and A. Zanchetti Cardiac and renal damage in the elderly hypertensive Journal of Renin-Angiotensin-Aldosterone System, March 1, 2002; 3(1_suppl): S16 - S24. [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||