[ Next ] [ Prev ] [ Abs ] [ Chi ] [ TOC ] [ Home ]
Chin Med J (Taipei) 1997;59:225-31.
Neurological Institute, Veterans General Hospital-Taipei and National Yang-Ming University School of Medicine, Taipei, Taiwan, R.O.C.
Background. Lacunar infarction (LI) is an ischemic stroke subtype with unique clinical, radiological and pathological features. Its relation to other stroke subtypes is unclear. To better understand the underlying pathological process of LI, we compared the risk factors of LI with those of other stroke subtypes.
Methods. During the study period (from January 1, 1990 to December 31,1991), 240 consecutive patients with first-ever strokes admitted to the stroke unit of our hospital were enrolled to the study and were classified into one of the four stroke subtypes (52 with LI, 80 atherothrombotic infarcts, 38 cardiogenic embolism and 70 brain hemorrhage) based on their computed tomography (CT) and clinical features using the guideline developed by the National Institute of Neurological Disorders. Eighty outpatients of similar age who had either low back pain or cervical spondylosis were recruited from the clinics of Neurology to serve as non-stroke controls. Data collected included demo-graphics, lifestyle, and other vascular risk factors. Detailed physical and neurological examination, blood biochemistry and Doppler ultrasound on cervical vessels were performed.
Results. Our investigations revealed that LI is a common stroke subtype accounting for 21% of all first-ever strokes in our hospital. Like ischemic stroke patients, those with LI were much more likely to have hypertension, diabetes, heart disease and carotid disease when compared with non-stroke controls. Patients with brain hemorrhage had less history of diabetes and lower levels of cholesterol than LI patients.
Conclusions. LI patients seemed to share more risk factors with ischemic stroke patients than with brain hemorrhage patients. These shared risk factors suggest a possibly similar underlying pathological process between ischemic strokes and LI patients. Careful screening for those risk factors should be part of the mandatory clinical management for the prevention of LI.
[Chin Med J (Taipei) 1997;59:225-31.]
Keywords: brain hemorrhage, ischemia, lacunar infarcts, risk factor, stroke
Received: July 30, 1996.
Accepted: January 17, 1997.
Address reprint requests to: Li-Chi Hsu, MD, Neurological Institute, Veterans General Hospital-Taipei, No. 201, Sec. 2, Shih-Pai Road, Taipei, Taiwan, R.O.C.
Ischemic strokes account for approximately eighty percent of all first-ever strokes [1,2]. The underlying pathogenesis varies among different subtypes of cerebral infarctions [3]. Lacunar infarction (LI) is a unique subtype and requires specific clinical and radiological features for diagnosis [4]. It is thought to result from the occlusion of a single perforating artery and might manifest clinically as a number of specific syndromes, namely, lacunar syndromes [5,6].
Hypertension has been mentioned as an important risk factor of LI, being present in approximately two thirds of LI patients [7-9]. However, since a substantial number of LI cases do not have hypertension, the role of hypertension in the pathogenesis of LI is questioned [10]. Other underlying causes such as embolism or hemodynamic compromise from a carotid lesion [4,8,9,11-13], or emboli from a cardiac source, have been postulated [4,9,14].
Although pathologic studies are of great value in the understanding of small vessel pathology leading to LI, the low case-fatality and the low autopsy rates in persons with LI preclude the feasibility of such studies. Therefore, the question of whether the underlying small vessel vasculopathy of LI is qualitatively different from those infarctions involving the cortical distribution of the major cerebral arteries or whether they are from the same entities remains unclear. Alternatively, spontaneous brain hemorrhage (BH) due to ruptures of same small penetrating arteries of LI is considered as leading to the theory of shared pathogenesis of these two stroke subtypes.
While a direct pathological study is not feasible, an indirect method may help to understand the underlying pathogenesis of LI. We sought to evaluate the association between some selected vascular risk factors (i.e. hypertension, heart disease and diabetes) and different stroke subtypes to provide indirect information for the underlying small vessel vasculopathy of LI. The purpose of this study was to compare the common vascular risk factors among different stroke subtypes. In addition, the differences of lipid profiles were evaluated.
The study was conducted in a tertiary care, university-affiliated, teaching hospital. From January 1, 1990 to December 31, 1991, 240 consecutive patients with first-ever strokes admitted to the stroke unit were enrolled. The study protocol was approved by the Institutional Review Board (IRB) of the study hospital. Informed consent was obtained through either the patients (if they were conscious) or their proxy (the closest relative).
Two neurologists independently reviewed each of the patient's clinical signs and symptoms as well as the computed tomography (CT) scan findings, and classified the patients in one of the four stroke subgroups following the guideline developed by the National Institute of Neurological Disorders and stroke [15]. When disagreement occurred between two reviewers, the final classification of that patient was determined by the consensus of the study panel (all the members of the research team). The four stroke subtypes of study interest were: LI, BH, atherothrombotic infarcts (AI), and cardiogenic embolism (CE). Patients who had a previously documented stroke or in hospital stroke were excluded.
A group of non-stroke patients (n=80) with matched age who had either low back pains or cervical spondylosis without any history of cerebrovascular disease and presented in our neurology outpatient clinic during the study period were recruited to serve as the control group.
Data were collected using a structured questionnaire and included sociodemographics, history of hypertension, diabetes mellitus, and other vascular risk factors. Detailed and standardized physical as well as neurological examinations were performed for all of the study subjects. After an 8-hour overnight fast, a venous blood sample was drawn from each of the study subjects for the measurement of serum blood glucose, triglyceride, cholesterol, blood urea nitrogen (BUN), creatinine, low density lipoprotein cholesterol (LDL-C) and high density lipoprotein cholesterol (HDL-C). Doppler scan of cervical vessels was also performed in all participating people using 5-MHz continuous-wave Doppler (Dopscan model 1050, Carolina Medical Electronics, King NC, USA) and/ or duplex scan with 5-MHz for Doppler and 7.5-MHz for imaging (HP SONOS 1000, Hewlett Packard, CA, USA). The diagnosis of carotid stenosis was based on Spencer's criteria [16], which has been reported to have high sensitivity and specificity [17].
Hypertension was considered to be present if blood pressure measurement was greater than or equal to 160/95 mmHg on two isolated occasions and required therapy after the acute stroke period (7 days after ictus) or if the patient was on chronic anti-hypertensive pharmaceutical therapy for previously diagnosed hypertension. Diabetes mellitus was considered present if there was a past history and hypoglycemic agent therapy was required. Heart disease was considered present if there was a significant past history of ischemic heart disease or arrhythmia such as atrial fibrillation. A smoker was defined as a person who smoked at least one cigarette per day for the preceding 3 months or more. The results of biochemical tests were considered abnormal if the numerical values were out of the normal range of our hospital's laboratory.
Data were first analyzed using descriptive statistics for frequency distributions and the group difference among categorical variables. Multivariate analysis was performed using logistic regression. Odds ratios and test-based 95% confidence intervals (95% CI) were calculated using the non-stroke control group or the LI group patients for reference.
Of the 240 stroke patients, 52 were classified in the LI group, 80 AI, 38 CE and 70 BH. The non-stroke control group patients were younger at the mean age than the four stroke subtype group patients. Table 1 lists the detailed sociodemographics.
Table 2 shows the results of multivariate analysis for each of the stroke subtypes using the non-stroke control group as a reference. Hypertension was an independent risk factor of both ischemic and hemorrhagic strokes. A history of heart disease was a risk factor of ischemic strokes, and was most prominent in patients with CE. Diabetes mellitus was a strong risk factor of both AI and LI. The frequencies of abnormal cholesterol and triglyceride levels did not differ between the controls and any of the four stroke subtypes. The percentage of low HDL-C level was significantly higher among patients with AI than in other study groups.
Table 3 shows the comparison of the risk factors of LI and other ischemic strokes. Odds ratios and 95% CI were calculated using the LI group as a reference. There was no significant difference in any of the risk factors listed in table 3 except higher frequency of heart disease and lower frequency of diabetes mellitus in the group of patients with CE.
Table 4 shows the comparison of the risk factors of LI and BH. Only the difference in a history of diabetes and the level of cholesterol reached the statistical significance. Although not statistically significant, there was a trend toward significance among other risk factors, such as heart disease, triglyceride level and carotid disease.
This two-year prospective study revealed LI as a common stroke subtype accounting for 21% (52/240) of all first-ever strokes in our hospital. Identified independent risk factors of LI were hypertension, heart disease and diabetes mellitus (Table 2). Our study demonstrated that patients with LI shared risk factors with other ischemic stroke patients, suggesting a possible common underlying vascular pathogenesis. Patients with BH were less likely to have history of diabetes mellitus and tended to have a low level of cholesterol compared with patients with LI, suggesting heterogeneity between LI and BH.
Our results were consistent with others' findings that hypertension was the most common risk factor of stroke subtypes [5]. For years, hypertension has been presumed the most common risk factor of LI [18-20]. Theoretically, the term "lacunar infarction" accepts the bipartite hypothesis that a small number of distinct clinical syndromes are associated with small deep infarcts or lacunes and that these lacunes are the result of occlusion of single perforating arteries primarily due to hypertension, the so called "lacunar hypothesis" [6]. In our study, patients with hypertension were 5.9 times more likely to incur LI compared with patients who did not have hypertension. This figure is compatible with previous reports [21,22]. However, a high odds ratio was also found among other stroke subtypes relative to the non-stroke patients in our study, indicating that hypertension was a shared risk factor of strokes regardless of subtype. In their community-based study of first-ever stroke, Lodder et al. [23] compared LI with carotid artery-distribution infarct involving the cortex and found no difference in the prevalence of hypertension between both groups, supporting our findings that hypertension contributes equally to both large and small artery diseases.
Although the role of heart in the pathogenesis of stroke has long been emphasized [24], it has only recently been recognized that heart disease is a frequent source of emboli to the brain [25,26]. Given the fact that not all patients with LI are hypertensive, cardiogenic emboli might be part of an important pathogenesis that we should not ignore. Fisher [27] found in two cases at autopsy that embolism seems the cause of lacunar infarction since there was no obstruction of the artery supplying the infarcted area. Others suggested that embolism from a cardiac source could result in LI in a subgroup of LI patients [28]. However, this issue is still debatable and inconclusive. You et al. [22] in their study of 203 patients with LI found no association between heart disease and LI . In our study, significantly more patients with LI than in the control group had a history of heart disease. Among our patients with LI, 17.3% (9/52) had a history of heart disease. This figure is similar to that in the patients with AI (15/80 = 18.8%) and much lower than in the patients with CE (22/38 = 57.9%), which was consistent with previous reports [12,28]. Given the definition of CE, not surprisingly, patients with CE were more likely to have a history of heart disease. Thus, our study suggests a possible pathophysiological role of cardiac embolism in a minority of LI patients.
Diabetes mellitus is a definite independent risk factor of atherosclerosis and atherothrombotic brain infarction [2,29,30]. It may also cause microatheroma in small vessels, such as lenticulostriate arteries, leading to LI. Our data revealed that diabetes was more prevalent in patients with either thrombotic or lacunar infarctions than in controls.
Hypercholesterolemia and hypertriglyceridemia were thought to be independent risk factors of arterial atherothrombosis . However, the relation between serum lipids and lipoproteins in cerebrovascular disease is not as clear-cut as in coronary atherosclerosis [31]. The present study failed to find any significant difference in the levels of cholesterol or triglyceride among various subgroups of stroke and controls. This result is consistent with the equivocal results of previous studies on the relevance of dyslipidemia with the incidence of ischemic stroke [32].
Carotid artery stenosis, especially the high degree stenosis (> 50% stenosis), carries a high risk for ischemic stroke [33-35]. Artery-to-artery embolism or hemodynamic compromise from a carotid lesion were thought to be rare causes of LI. [4,8,9,11-13,28]. According to our data, greater-than-50% carotid stenosis was more prevalent in patients with ischemic strokes than in controls, but there was no difference among the 3 subtypes of ischemic stroke.
BH is more common among people in Taiwan than in other countries [36]. It most commonly results from rupture of microaneurysm caused by angionecrosis (arteriolonecrosis, fibrinoid necrosis, or lipohyalinosis) in intracerebral penetrating arteries [37,38]. Some investigators have found lower serum total cholesterol level in patients suffering from BH [39,40]. Low serum cholesterol is supposed to contribute to the development of a fragile cerebrovascular endothelium, and the eventual development of angionecrosis and cerebral hemorrhage in the presence of hypertension [40]. However, our study found no difference in the level of total cholesterol among patients with ischemic strokes, BH and controls.
One unique aspect demonstrated in our study was that patients with BH were less likely to have diabetes mellitus and had lower level of cholesterol than patients with LI. Thus, differences do exist in the pathogenesis underlying both BH and LI, although they both come from small vessel vasculopathy. There seems no proper explanation for less cases of diabetes mellitus among patients with BH. Probably DM does not contribute in any significant way to the occurrence of small vessel lipohyalinosis that underlies BH, but its elucidation is awaiting further studies.
In conclusion, LI shares many vascular risk factors with other ischemic strokes, suggesting a possible common underlying vascular pathogenesis. Careful screening for the risk factors could help preventing LI and should be part of mandatory clinical management. However, other heterogeneous etiologic factors may contribute to the pathogenesis process of LI as a distinct form of ischemic stroke. Searching for other risk factors thoroughly in a patient presenting with LI is warranted.
This study was supported in part by Tjing Lin Medical Foundation.
Copyright: 1997, Chinese Medical Association (Taipei)