Frequency of increased (>0%), reduced (−10% to 0%), and normal (<−10%) systolic circadian blood pressure variation in patients with lacunar infarction and control patients. Asterisk indicates P<.05.
Odds ratios with 95% confidence intervals for having lacunar infarction related to blood pressure and systolic circadian blood pressure variation. Odds ratio=1.0, defined for normotensive patients (diastolic daytime average blood pressure of ≤85 mm Hg) with systolic circadian blood pressure variation less than −5 %. A, Normotensive with a systolic circadian blood pressure variation less than −5%; B, normotensive with a systolic circadian blood pressure variation greater than −5%; C, hypertensive with a systolic circadian blood pressure variation less than −5%; and D, hypertensive with a systolic circadian blood pressure variation greater than −5%.
Kukla C, Sander D, Schwarze J, Wittich I, Klingelhöfer J. Changes of Circadian Blood Pressure Patterns Are Associated With the Occurrence of Lacunar Infarction. Arch Neurol. 1998;55(5):683-688. doi:10.1001/archneur.55.5.683
The occurrence of lacunar infarction is closely related to arterial hypertension. However, there is only limited and partly controversial knowledge regarding the possible pathogenetic role of circadian blood pressure changes.
To evaluate the relationship between circadian blood pressure rhythm, occurrence, and extent of lacunar infarction.
We analyzed circadian blood pressure patterns, other cardiovascular risk factors, and occurrence of lacunar infarction in 118 hospitalized patients older than 55 years. Noninvasive 24-hour blood pressure measurements and magnetic resonance or computed tomographic brain imaging were performed in 61 patients with lacunar infarction and in 57 control patients. Daytime blood pressure variability was defined as the within-subject SD of all systolic and diastolic blood pressure readings during the daytime measurement period. Circadian blood pressure variation was defined as the average percentage change of nighttime blood pressure values compared with the daytime blood pressure values.
Patients with lacunar infarction were significantly older and showed more often a history of arterial hypertension, elevated average daytime blood pressure values, an increased systolic daytime blood pressure variability, and a reduced circadian blood pressure variation due to an increased incidence of a pathologic nighttime blood pressure increase. No significant correlation was found between these parameters and the number of lacunae. A logistic regression analysis revealed that a reduced systolic circadian blood pressure variation, age, systolic average daytime blood pressure, and a history of arterial hypertension were best correlated with the occurrence of lacunar infarction.
Reduced nighttime decline in systolic blood pressure may be an important risk factor for the development of lacunar infarction in addition to the absolute level of blood pressure and age.
LACUNAE ARE small, deep infarcts due to penetrating arteriolar occlusive disease and can produce characteristic lacunar stroke syndrome.1 With increasing use of magnetic resonance (MR) and computed tomographic (CT) scans, clinically silent lacunar infarction has been shown to occur frequently among elderly patients.2,3 In addition to age and other cardiovascular risk factors, lacunar infarction is closely related to arterial hypertension.4 It has recently become possible to determine 24-hour blood pressure patterns using a noninvasive blood pressure monitoring device. Several studies have shown that target-organ damage correlates more closely with 24-hour average blood pressure and 24-hour blood pressure variability than with casual blood pressure.5- 7
However, there is only limited knowledge regarding the possible pathogenetic role of circadian blood pressure changes for the occurrence of lacunar infarction. In normotensive subjects and in most of those with primary hypertension, the systolic and diastolic average nighttime values were approximately 15% below the daytime values ("dipper").8- 10 In some forms of secondary hypertension, but also in primary hypertension with obesity11 or with left ventricular hypertrophy,12- 14 the characteristic nighttime blood pressure decrease is reduced or missed completely ("nondipper").9 In patients with lacunar infarction, a reduced nighttime blood pressure decrease was inconsistently found in previous studies.15- 17
Furthermore, it is still debatable whether there are differences in blood pressure characteristics in single symptomatic lacunar infarctions and multiple asymptomatic lacunar infarctions.18 Knowledge about pathologic changes of circadian blood pressure rhythmicity may be useful in prevention and in medical treatment of patients with lacunar infarction. The objective of our study was therefore to analyze the changes of circadian blood pressure patterns in patients with lacunar infarction and their relation to the extent of lacunar infarction.
From a series of 230 consecutive inpatients older than 55 years with brain MR or CT imaging results and hospitalized for acute neurologic disorders, 118 patients were included in the present study (mean age, 70 years; 95% confidence interval [CI], 68-71; 49 women and 69 men). Patients were excluded from the study if they had cerebral infarction (lacunar and nonlacunar strokes) up to 6 months before the present study (n=37), had only possible lacunar lesions or only leukoaraiosis (n=62), were taking cortisol medication (n=7), or had insufficient blood pressure measurements (n=6). All patients underwent MR (n=33) or CT (n=85) imaging. Based on the results of the imaging study, the patients were subdivided into 2 groups: 61 (26%) of the 230 consecutively studied patients showed asymptomatic lacunar infarction and 57 patients showed no abnormality. The patients with a normal brain scan were used as a control group. All patients were under our medical jurisdiction during their clinical course. There were no significant differences between the control group and the lacunar infarction group regarding type and severity of the patients' diseases. Bedridden patients were excluded. The essential data of the 2 groups are shown in Table 1. The study was approved by the local institutional review board. All patients provided informed consent before entering the study. Furthermore, we determined a history of arterial hypertension (systolic blood pressure of >160 mm Hg and/or diastolic blood pressure of >95 mm Hg by conventional measurements8,19), smoking (during the last 5 years), diabetes mellitus, cholesterol levels, and triglyceride levels at admission in all patients.
Noninvasive 24-hour blood pressure measurements were performed in all patients during their clinical course (mean, 4.9 days after admission to hospital; 95% CI, 4.1-5.7 days) using an automatic portable blood pressure monitor with an oscillometric measurement device (Spacelabs ABD-Monitor 90207, Spacelabs GmbH, Kaarst, Germany). The noninvasive measurements were obtained with an adult-size cuff (2 sizes available) and at a cuff deflation rate of 4 mm Hg per second. No significant difference could be determined by comparing the predecessor monitor (SL 90202) with intra-arterial measurements.20 In general, blood pressure values on admission were more frequently elevated in both patient groups. However, to exclude a "white-coat" effect and because there were no patients with acute cerebral infarction, we performed the 24-hour blood pressure measurements at least 48 hours after admission. We did not perform repeated monitoring regularly, because the reliability and validity of the blood pressure monitor used in this study has been demonstrated in previous studies.20,21 All blood pressure measurements were done on an inpatient basis with similar daily activities. The measurements were made at intervals of 15 minutes during the entire 24-hour period (mean, 74.4 successful measurements per patient; 95% CI, 72.0-76.8). All measurements marked with an error event code on the monitor that were caused by movement artifacts were manually omitted from the calculations. The systolic, diastolic, mean blood pressures, and heart rates were reported. According to the recommendations of several population-based studies,10,22 the average daytime values were determined between 6 AM and 10 PM and the average nighttime values were determined between 10 PM and 6 AM. Daytime blood pressure variability was defined as the within-subject SD of all systolic and diastolic blood pressure readings during the daytime measurement period. Circadian blood pressure variation was defined as the average percentage change of nighttime blood pressure compared with the daytime blood pressure values. Eight patients interrupted the nighttime blood pressure measurements and these values were omitted from calculations of circadian blood pressure variation. Based on the results of the initial 24-hour blood pressure measurements, we defined arterial hypertension (diastolic average daytime blood pressure of >85 mm Hg) and isolated systolic hypertension (systolic daytime average blood pressure of >135 mm Hg and diastolic daytime average pressure of <85 mm Hg).10,22 Initiation or changes of antihypertensive medication were not performed during the 24-hour measurement. We only analyzed the initial long-term blood pressure measurement, because afterward antihypertensive therapy was optimized in patients suffering from arterial hypertension. All patients with a history of hypertension and 6 patients without a history of hypertension received antihypertensive medication during the 24-hour measurement period (33% in the control group vs 56% in the lacunar infarction group). Patients taking cortisol medication were excluded (n=7), because cortisol might affect circadian blood pressure rhythmicity. During the measurement period, none of the patients used sedatives. Sleep disturbances were comparable between patients with lacunar infarction and control patients. The long-term measurement of blood pressure was made on the left side in right-handed patients and vice versa after relevant differences between the sides had been ruled out by conventional checks of blood pressure and the proper positioning of the cuff was ensured in each subject. The patients were instructed to keep their arms as still as possible during the measurement period. All patients kept a number-coded diary in which activities and particular events were recorded, specifying the respective time. To ensure consistent measurement conditions and to minimize measurement mistakes caused by movement artifacts, patients were instructed to avoid intensive physical activity during the measurement period.
Evaluation of lacunar infarction was performed using MR (n=17) (Philips Gyroscan ACS-II 1.5 T, Philips GmbH, Hamburg, Germany) or CT (n=44) (Siemens Somatom DR-H, Siemens AG, Erlangen, Germany) imaging. Lacunae were defined as low-signal intensity areas (<1.5 cm in greatest diameter) on CT and T1-weighted MR images, which were visible as hyperintense lesions on T2- and proton density–weighted MR images. Lacunae were counted only if they appeared on both (short and long repetition time) spin-echo pulse sequences. Patients with possible lacunar lesions, visible only on the T2-weighted MR images,3 were excluded. A neuroradiologist, unaware of the clinical data, evaluated the number of lacunae from the MRI and CT data. Lacunar infarcts were located in the basal ganglia (58%), white matter (19%), cerebellum (11%), and brainstem (12%).
All values are given as mean and 95% CI. Independent t tests and Fisher exact tests were used to test differences between both groups. To evaluate the effect of the different risk factors on lacunar infarction, a multiple logistic regression analysis was performed using forward selection followed by backward elimination of covariates, resulting in an equation in which only covariates that significantly increase the predictability of the dependent variable are included. All covariates included in the final model were tested for interactions with each other. Because the tolerance values for each covariate were above 0.5, no correction for collinearity of the data was necessary. Age, smoking (no=0, yes=1), diabetes mellitus (no=0, yes=1), history of arterial hypertension (no=0, yes=1), cholesterol levels, triglyceride levels, systolic and diastolic daytime blood pressure values, systolic and diastolic circadian blood pressure variation, and daytime systolic and diastolic blood pressure variability were selected as independent variables; the occurrence and number of lacunae were the dependent variables. The statistical calculations were performed using a computerized statistical program (SYSTAT, Systat Inc, Evanston, Ill). A calculated difference of P<.05 was considered statistically significant.
Patients with lacunar infarction were significantly older and showed more often a history of arterial hypertension, elevated average daytime blood pressure values, an increased systolic daytime blood pressure variability, and a reduced circadian blood pressure variation (Table 1). A pathologic increase of systolic nighttime blood pressure (>0%) was significantly more frequent in patients with lacunar infarction than in the control patients (23% vs 7%; P<.05) (Figure 1). In contrast, no significant differences for diastolic nighttime blood pressure increase (14% vs 7%; P=.36) and decline of nighttime heart rate were found (Table 1). Patients with lacunar infarction showed significantly more often a systolic blood pressure variability greater than 15 mm Hg compared with the control group (49% vs 28%; P<.05). No significant differences were observed for all patients and those with lacunar infarctions between normotensive and hypertensive patients regarding the circadian blood pressure variation. Compared with normotensive control patients with normal systolic circadian blood pressure variation (<−5%; odds ratio, 1.0), lacunar disease was significantly more frequent in hypertensive subjects with normal circadian blood pressure variation (60% vs 31%; odds ratio, 3.2; 95% CI, 1.3-8.2). Patients with both hypertension and reduced systolic circadian blood pressure variation (>−5%) demonstrated the highest frequency of lacunar infarction (83%) (odds ratio, 10.8; 95% CI, 2.6-44.6; Figure 2). No significant difference was observed between patients with single (n=28) or multiple (3 or more, n=17) lacunar infarctions with regard to age, cardiovascular risk factors, and blood pressure parameters.
To evaluate the influence of the different risk factors (age, smoking, diabetes mellitus, history of hypertension, cholesterol levels, triglyceride levels, systolic and diastolic blood pressure, systolic and diastolic circadian blood pressure variation, and daytime systolic and diastolic blood pressure variability) on occurrence and extent of lacunar infarction, a stepwise multiple regression analysis was performed. Whereas no significant relation with the number of lacunae was found, the average daytime systolic blood pressure, systolic circadian blood pressure variation, age, and history of hypertension were significantly related to the occurrence of lacunar infarction (Table 2). All other tested risk factors did not significantly increase the predictability of the regression. The predicted model accounted jointly for 29% of the patients with lacunar infarction.
Lacunar infarction was found in 26% of our patients, with a mean number of 2.1 lacunae. This frequency was slightly higher than that reported in previous studies evaluating incidence rates of lacunar infarction on CT brain imaging in hospitalized patients (14%23) or autopsy studies (6%24 and 8%1). Data concerning the incidence of asymptomatic lacunar infarction on MR brain imaging in healthy elderly subjects ranged from 9%3 to 47%2 probably due to the age differences in the 2 study populations. Magnetic resonance and CT imaging are valid instruments for the visualization of lacunar infarction, in which MR imaging is thought to be more sensitive in detecting very small lesions.25,26
Our results confirm the importance of arterial hypertension as a major risk factor for lacunar infarction.4,27 In addition to arterial hypertension, we found that the occurrence of lacunar infarction was closely related to a reduced circadian blood pressure variation in normotensive and hypertensive subjects. Former studies have linked nondippers with an increased incidence of cardiovascular28 and cerebrovascular morbidity.15,29 However, findings regarding the percentage change of nighttime blood pressure compared with the daytime blood pressure values in patients with lacunar infarction are controversial and limited because many investigators analyzed the absolute amplitude of nocturnal blood pressure change2,15,17 and the absolute nocturnal blood pressure values.2 Previous studies using noninvasive 24-hour blood pressure measurements have shown that nighttime blood pressure values, particularly during sleep, correlate with the number of lacunae,2 and were higher in patients with multiple lacunar infarctions compared with control patients.16 Reduction of circadian blood pressure variation using the absolute blood pressure amplitude was reported in patients with vascular dementia of the Binswanger type, patients with dementia of the lacunar type, and even in single lacunar infarction.15 However, other investigators observed that lacunar infarction was not significantly related to a reduced absolute amplitude of nighttime blood pressure decline.2 We could demonstrate that lacunar infarction is related to reduced circadian blood pressure variation using a definition, which gives more independence from absolute blood pressure values. However, one could argue that a different nighttime blood pressure decrease between normotensive and hypertensive patients has to been taken into account when interpreting the differences of circadian blood pressure variation averaged over all patients. In persons with essential hypertension, the nighttime decrease of blood pressure is comparable to that of, or even more pronounced than, normotensive subjects.30 The significant reduction of circadian blood pressure variation in the lacunar infarction group despite the increased frequency of essential hypertension points to a disturbed circadian rhythmicity in patients with lacunar infarction. In a previous study, using also the percentage of nighttime blood pressure decline, patients with multiple lacunae showed reduced circadian variation compared with control patients.16 In contrast to our findings and that of others,2,15,16 a recent report observed that an excessive decrease in nighttime blood pressure was related to lacunar infarction in elderly Japanese women, but not in men.17 The observed differences between that study and our results and those of others appear attributable to differences in the patient's age, the only univariate analysis and the outpatient population in the study of Watanabe et al.17
One could argue that our results are a consequence of infarction.16 A reduction of circadian blood pressure variation following stroke has been described previously.31,32 We excluded patients with a history of infarction 6 months before the present investigation, whereas others did not.15,16 Heart rate reduction during the night seems to be a valuable parameter to distinguish nondippers with acute ischemic stroke and nonsleepers from nondippers due to, for instance, long-standing essential hypertension.33,34 In our study, we found no significant change in nighttime heart rate decrease in patients with lacunar infarction compared with the control patients. However, our present study is cross-sectional and, therefore, we cannot completely exclude the possibility that a nondipper profile is a persisting result from lacunar infarction as described by Kario and Shimada35 in a case report.
As in any hospital-based study, this one may also suffer from referral bias. One could argue that our results may be influenced by a selection bias concerning the patient population because blood pressure measurements were performed on an inpatient basis and in patients with neurologic disorders. We excluded bedridden patients, and in none of the patients was there a secondary cause of hypertension. In addition, we ensured that patients with more serious diseases were not more frequently represented in one group than in the other. Thus, we believed that a possible selection bias was only of minor importance according to our main results.
We observed that an elevated systolic circadian blood pressure variability is related to the occurrence of lacunar infarction in an univariate analysis. This finding is in accordance with other studies suggesting that target-organ damage is greater and incidence of cardiovascular morbid complications is higher in hypertensive people with high blood pressure variability than in those with low variability.5,36 Furthermore, systolic daytime blood pressure variability was found to be the best predictor of early atherosclerosis of the common carotid artery.37 In contrast to its important role in large-vessel disease, our results using multivariate regression analysis, however, revealed that an increased blood pressure variability is of only minor importance for the development of lacunar infarction.
Surprisingly, we found no relation between blood pressure parameters and other risk factors on extent of lacunar infarction either in a univariate or multivariate analysis. Furthermore, there were no significant differences in patients with single lacunar infarction compared with those with multiple lacunar infarctions. A previous study18 reported hypertension and leukoaraiosis might be more closely related to multiple lacunar infarction than to a single, symptomatic lacunar infarct. They found that patients with asymptomatic, multiple lacunar infarcts had hypertension significantly more often than those patients with only a symptomatic lacunar infarct. Using multivariate logistic regression analysis, however, leukoaraiosis, but not hypertension was a significant and independent predictor of multiple asymptomatic lacunar infarction.18 We suggest that a combined effect of risk factors in addition to age, hypertension, and pathologic changes of circadian rhythmicity might cause multiple lacunar infarction.
From a clinical point of view our findings are consistent with the results of others, warranting medical treatment even of isolated systolic hypertension.19,38- 40 The occurrence of lacunar infarction was strongly associated with changes of systolic circadian blood presssure patterns but not with diastolic blood pressure patterns. Contrary to earlier beliefs that the diastolic component of blood pressure is the major determinant of cardiovascular risk, there is in accordance to our results a growing body of evidence that systolic hypertension may be a more important contributor to risk than its diastolic counterpart.19 Borderline isolated systolic hypertension was found to be the most common type of untreated hypertension among adults older than 60 years in the Framingham Heart Study and was associated with increased risk of progression to definite hypertension and the development of cardiovascular disease. In the light of these findings, earlier interventions would appear justified.19,38- 40 In addition, demonstration of lacunar infarction should lead to a careful evaluation of nighttime blood pressure increase even in the absence of daytime arterial hypertension. If a reduced nighttime decline in blood pressure and particularly an increase is present, antihypertensive treatment in the evening or a long-acting agent may be useful to normalize nighttime blood pressure and thus may decrease the risk of further cerebrovascular complications. As standard stroke treatment seems to fail in stopping the progression of small, deep infarct,41 new therapeutic strategies are even more necessary.
Accepted for publication September 19, 1997.
Supported by the Stiftung Deutsche Schlaganfall-Hilfe, Gütersloh, Germany.
We would like to thank Helga Gräfin von Einsiedel, MD, Department of Neuroradiology, Technical University of Munich, Munich, Germany, for her helpful support during the evaluation of the magnetic resonance imaging and computed tomographic data.
Reprints: Jürgen Klingelhöfer, MD, Department of Neurology, Technical University of Munich, Moehlstrasse 28, 81675 München, Germany.