- Email: [email protected]
Correlation of retinopathy with leukoaraiosis in patients with anterior circulation infarcts
Journal of Clinical Neuroscience xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Clinical Study
Correlation of retinopathy with leukoaraiosis in patients with anterior circulation infarcts Wei Wei a, Zhenxi Xia a, Hongyan Gao b, Jingjing Gong a, Liping Yan c, Yonghua Huang a,⇑, Fei Chen a, Weiwei Zhang a a b c
Department of Neurology, PLA Army General Hospital, No.5 Nan men Cang, Dongcheng District, Beijing 100700, China Department of Medical Affairs, General Hospital of Beijing Command PLA, Dongcheng District, Beijing, China Department of Neurology, TangshanWorker Hospital, Tangshan, China
a r t i c l e
i n f o
Article history: Received 20 November 2015 Accepted 14 February 2016 Available online xxxx Keywords: Anterior circulation infarcts Brain microvascular diseases Leukoaraiosis Retinopathy Risk factors
a b s t r a c t Although retinal and cerebral microvessels share similar embryological, anatomical and physiological characteristics, the correlation between retinopathy and leukoaraiosis (LA), a type of brain microvascular disease, is unclear. In the present study, the sample included 213 patients admitted to the department of neurology from January 2012 through October 2012. MRI and retinal photography were performed within 48 hours of hospitalization, and patient demographics, comorbidities, preadmission medications and laboratory data were collected. MRI images were used to divide the patients into LA and non-LA groups. Using multivariate binary logistic regression, the effects of retinopathy on LA were investigated. Of the 213 patients enrolled, 168 were included in this study (LA, n = 108; non-LA, n = 60). Hypertension, coronary heart disease and carotid artery plaque were more common in the LA group, and these patients showed higher blood levels of C-reactive protein, homocysteine and triglycerides. The incidence of retinopathy was significantly increased in the LA group compared with the non-LA group, and there was a significant correlation between the severity of LA and incidence of retinopathy. Retinopathy is an independent risk factor for LA and can significantly increase the risk of LA when combined with age, coronary heart disease, C-reactive protein, carotid artery plaque or systolic pressure. Taken together, retinopathy is associated with LA in patients with anterior circulation infarcts. Retinopathy is an independent risk factor for LA and an increase the risk of LA, and thus facilitating the evaluation of LA. Ó 2016 Published by Elsevier Ltd.
1. Introduction Leukoaraiosis (LA), defined as a rarefaction or ‘‘araiosis” of the brain white matter, was first proposed by Hachinski in 1987 as a neuroimaging diagnostic term [1]. It is showing up as decreased signal density on CT scan and increased density on T2-wighted MRI/fluid attenuated inversion recovery (FLAIR) MRI sequences performed as MRI scans [2,3]. It mainly appears in the subcortical white matter, centrum semiovale, periventricular and/or deep white matter, where they receive the blood supply primarily from the terminal arteries of the middle cerebral artery and the terminal branches of the choroidal artery. It is well recognized that LA is a type of brain microvascular disease with important clinical implications [4]. Accumulating evidences have demonstrated that the LA commonly occurs in stroke patients and closely associated with the occurrence and clinical outcome of both ischemic [1,5,6] and ⇑ Corresponding author. Tel.: +86 010 84008399; fax: +86 010 64056642. E-mail address: [email protected] (Y. Huang).
hemorrhagic [7,8] stroke. As a composite marker of baseline brain integrity, the presence or severity of LA is also strongly associated with the risk of vascular dementia [4,9] and Alzheimer’s disease [10,11]. Retinal and cerebral microvessels share a common embryological origin, anatomical basis and physiological characteristics [12,13]. A postmortem study of stroke patients have revealed that the pathological alterations of retinal vessels are similar to those of cerebral microvessels [14]. Incomplete white matter infarction caused by stenosis or occlusion of small cerebral vessels with sudden or chronic ischemia is considered to play a crucial role in the pathogenesis of LA [15]. LA progression over time is substantial, especially in patients with certain phenotypes [16]. During LA progression in cerebral microvessels, the stenosis or occlusion may also occur in the retina as there is a close relationship between the microvessels in the eye and the brain. Many retinal conditions, including hypertensive retinopathy, diabetic retinopathy, and retinal arteriolar emboli, have been reported to be associated with stroke, reflecting concomitant pathophysiological processes
http://dx.doi.org/10.1016/j.jocn.2016.02.042 0967-5868/Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Wei W et al. Correlation of retinopathy with leukoaraiosis in patients with anterior circulation infarcts. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2016.02.042
2
W. Wei et al. / Journal of Clinical Neuroscience xxx (2016) xxx–xxx
affecting both the eye and the brain [17]. However, the relationship between retinal conditions and LA has not been studied yet. In this study, a total of 213 anterior circulation infarcts patients were recruited and divided according to the presence of LA. MRI scans and retinal photography were performed within 48 hours of hospitalization, and patient demographics, comorbidities and laboratory data were collected. Image analysis of LA and the assessment of retinopathy were carried out by two experienced neurologists with high inter-rater agreement. By comparing retinopathy and other confounding factors in the LA and non-LA groups and applying multivariate logistic regression, the correlation between retinopathy and LA was investigated. 2. Methods 2.1. Patients In this study, we consecutively recruited 213 patients with anterior circulation infarcts who were admitted to our hospital from January 2012 to October 2012. All patients underwent MRI scans and retinal photography within 48 hours of admission. According to the presence of white matter hyperintensity on the FLAIR sequence and T2-weighted MRI, the subjects were divided into two groups: an LA group and a non-LA group. 2.2. Inclusion criteria The selected patients were in the age of 40–80 years and able to cooperate with the examiner to complete the MRI scan in the supine position or retinal photography in the seated position. 2.3. Exclusion criteria The patients were not eligible if they presented the following: (1) white matter disease caused by multiple sclerosis or infectious, metabolic, toxic, autoimmune and metastatic diseases; (2) history of hemorrhagic stroke, brain tumors or cerebral hemorrhage; (3) concomitant systematic disorder that potentially affects the optic fundus (for example, severe liver, kidney or heart failure; severe infection; malignant disease; systemic lupus erythematosus; or hereditary disease); (4) local eye disorders that could cause an optic fundus disease, such as various eye inflammatory responses or eye surgeries (e.g., cataract extraction or laser surgery) within six months prior to admission; (5) narrow anterior chamber angle; (6) allergy to mydriatic agents, or (7) disturbance of consciousness. 2.4. Collection of clinical data Patient demographics, comorbidities, preadmission medications and laboratory data were collected within 48 hours of hospitalization. Specifically, we assessed the presence of hypertension (at least two blood pressure readings of P140/90 mmHg on separate days measured after a 20-minute rest or taking antihypertensive medication), diabetes mellitus (at least two random serum glucose results >11.1 mmol/L or a fasting serum glucose level >7.0 mmol/L, a specialist’s diagnosis or the use of diabetic medication), coronary heart disease (a specialist’s diagnosis or the use of medication), smoking history (>1 cigarette per day for >1 year), and drinking history (100 g alcohol intake per day for >5 years), and the data were recorded based on the corresponding criteria. Lacunar infarction and atherosclerotic carotid plaque were documented after the MRI scan and ultrasonography, and basic biochemical indexes were collected, including fasting serum glucose, cholesterol, triglycerides, C-reactive protein (CRP) and homocysteine.
2.5. Neuroimaging protocol and image analysis of leukoaraiosis MRI scans were obtained from all patients using a 3.0 T SIEMENS MRI scanner, with sequences that included FLAIR and T1-weighted and T2-weighted gradient echo. The severity of LA was graded based on the periventricular white matter hyperintensity according to the Fazekas scale: 0, absent; 1, caps or pencil-thin lining; 2, smooth halo; or 3, irregular periventricular hyperintensity extending into the deep white matter. LA was reviewed independently by 2 experienced, board-certified neurologists who were blinded to both the research design and the clinical data. To evaluate the degree of inter-rater agreement, a Kappa statistic was calculated from the findings of 30 randomly chosen patients. The Kappa value was >0.80, indicating that the two neurologists had almost perfect consistency when determining the presence or extent of LA. 2.6. Retinal photography and assessment Retinal photography was conducted in all patients after pharmacological mydriasis by experienced ophthalmologists, and fundus color transparencies were taken (50° field centered on the optic disc), followed by an assessment of the presence of retinopathy by 2 trained graders (the neurologists mentioned above). The determination of retinal arteriovenous (AV) nicking, microaneurysms, cotton-wool spots, hard exudates, and retinal hemorrhage was based on standard protocols [18]. For quality control, a random subsample of 30 retinal photographs was examined and the assessment results from the 2 graders were tested using a weighted Kappa analysis. The Kappa values revealed that the inter-grader agreement was also high. 2.7. Standard protocol approval and patient consent The design and implementation of this research was approved and documented by the medical ethics committee at our hospital, and written informed consent was obtained from the patient or a legally authorized representative in all cases. 2.8. Statistics To evaluate the degree of inter-rater agreement, we examined weighted Kappa statistics in a cohort of 30 randomly chosen patients. According to the Kappa values, the inter-rater agreement was interpreted as slight (0.01–0.2), fair (0.21–0.4), moderate (0.41–0.6), substantial (0.61–0.8), or almost perfect (0.81–1.0). Continuous variables are presented as the mean ± standard deviation (SD), and categorical variables are reported as percentages. Continuous variables were assessed for a normal distribution using the one-sample Kolmogorov–Smirnov test. Normally distributed continuous variables were processed using independent-sample t-tests, and non-normally distributed continuous data were compared using the Mann–Whitney U test. Categorical variables were analyzed using the v2 test. To evaluate the correlation between LA and retinopathy, binary logistic regression was used, and an analysis of the risk factors for LA or retinopathy was also performed using multivariate logistic regression. Two-sided tests were used for all analyses, and P < 0.05 was defined as significant. All statistical analyses were performed using IBM SPSS version 18.0. 3. Results A total of 213 patients with anterior circulation infarcts were enrolled in this study. Among these patients, eight were excluded due to severe systematic diseases (three for severe heart failure,
Please cite this article in press as: Wei W et al. Correlation of retinopathy with leukoaraiosis in patients with anterior circulation infarcts. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2016.02.042
3
W. Wei et al. / Journal of Clinical Neuroscience xxx (2016) xxx–xxx
four for malignant disease, and one for systemic erythematosus), and 13 were excluded due to various levels of disturbed consciousness (n = 12) or disturbed consciousness in combination with posterior circulation infarcts (n = 1). Of the remaining 24 patients who were excluded, 14 did not undergo brain MRI within the first 24 hours of admission; three declined to participate in this study; and seven did not complete the retinal photography. The remaining 168 patients formed the final eligible population (n = 108 in the LA group and n = 60 in the non-LA group). The demographic and clinical characteristics of the LA and non-LA groups are shown in Table 1 and Figure 1. There were no differences in the sex ratio, smoking or alcohol history, nor in the incidence of diabetes between the LA and non-LA groups. Hypertension and carotid artery plaque were more frequently observed in the LA group compared with the non-LA group. The patients with LA had higher levels of C-reactive protein (P < 0.01), homocysteine (P < 0.01) and triglycerides (P < 0.04) in the blood and were more likely to have high blood pressure, both systolic (P < 0.01) and diastolic (P < 0.04) (Fig. 1). No significant differences were observed in fasting serum glucose, high-density lipoprotein, or low-density lipoprotein between the two groups.
The incidence of retinopathy was significantly increased in the LA group compared with the non-LA group (88.0% vs. 65.0%, p < 0.001). With the exception of cotton-wool spot (3.7% vs. 6.7%), most signs, including A-V nicking (70.4% vs. 38.3%), microaneurysm (11.1% vs. 5.0%), hard exudate (27.6% vs. 15.0%) and hemorrhage (7.4% vs. 5.0%), were more frequently observed in the LA group than in the non-LA group (Table 2). Among these signs, differences in the incidences of A-V nicking and hard exudate were significant (Table 2). Moreover, LA occurred more frequently in patients with any type of retinopathy than in patients without retinopathy (Fig. 2). Patients with any retinopathies, particularly microaneurysm (p = 0.002) and hard exudate (p < 0.001), exhibited more severe LA than patients without retinopathy (Fig. 3). There was a significant correlation between the severity of LA and the incidence of retinopathy when all 168 patients were analyzed together (r = 0.186, p = 0.016). The patients with retinopathy had a 3.935fold greater risk (95% confidence interval [CI]: 1.793–8.633, p < 0.001) of LA than those without retinopathy. After adjustment for the other risk factors of LA, the odds ratio still remained at 3.745 (95% CI: 1.100–12.752, p = 0.040, Table 3), indicating that retinopathy is an independent risk factor for LA.
Table 1 Demographic characteristics of leukoaraiosis and non-leukoaraiosis groups Variables Sex
Male, n (%) Female, n (%)
Age (years), mean ± SD Smoking history, n (%) Alcohol history, n (%) Hypertension, n (%) Diabetes mellitus, n (%) Coronary heart disease, n (%) Lacunar infarction, n (%) Carotid artery plaque, n (%)
LA (n = 108)
Non-LA (n = 60)
P value
53 (49.1) 55 (50.9) 65.55 ± 11.16 41 (38.0) 30 (27.8) 75 (68.8) 50 (46.5) 48 (44.3) 70 (64.8) 68 (57.7)
29 (48.3) 31 (51.7) 62.73 ± 9.37 22 (36.7) 17 (28.3) 32 (53.3) 18 (29.0) 19 (31.7) 29 (48.3) 30 (50.0)
0.88 0.71 0.10 0.86 0.67 0.04* 0.22 0.11 0.04* 0.10
* P < 0.05 was considered statistically significant. LA = leukoaraiosis.
Fig. 1. The clinical characteristics of patients in the leukoaraiosis and non-leukoaraiosis groups.
Table 2 Incidence of each type of retinopathy, n (%) Group
N
A-V nicking
Microaneurysm
Hard exudate
Cotton-wool spot
Hemorrhage
LA Non-LA P value
108 60
76 (70.4) 23 (38.3) 0.001*
12 (11.1) 9 (5.4) 0.18
34 (27.6) 9 (15.4) 0.019*
4 (3.7) 4 (8.0) 0.39
8 (7.4) 3 (5.0) 0.55
* P < 0.05 was considered statistically significant. LA = leukoaraiosis.
Please cite this article in press as: Wei W et al. Correlation of retinopathy with leukoaraiosis in patients with anterior circulation infarcts. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2016.02.042
4
W. Wei et al. / Journal of Clinical Neuroscience xxx (2016) xxx–xxx Table 3 Effects of retinopathy and other risk factors on LA (logistic regression) P value
Fig. 2. Percentage of patients with various leukoaraiosis grades for each subtype of retinopathy. All patients in the leukoaraiosis and non-leukoaraiosis groups were analyzed together.
Retinopathy Age Smoking history Alcohol history Hypertension Diabetes mellitus Coronary heart disease Carotid artery plaque Systolic pressure Diastolic pressure Fasting serum glucose Triglyceride Cholesterol High-density lipoprotein Low-density lipoprotein C-reactive protein Homocysteine
*
0.035 0.042* 0.297 0.651 0.316 0.849 0.564 0.983 0.499 0.161 0.466 0.287 0.704 0.931 0.674 <0.001* 0.105
OR
95% CI
3.745 2.843 1.990 0.752 0.605 1.107 1.318 0.989 1.012 1.039 0.896 1.451 0.793 1.093 1.298 1.391 2.290
1.100–12.752 0.993–1.094 0.546–7.254 0.219–2.586 0.227–1.615 0.387–3.173 0.515–3.371 0.351–2.788 0.977–1.048 0.985–1.097 0.668–1.203 0.732–2.878 0.240–2.619 0.145–8.262 0.385–4.376 1.213–1.595 0.841–6.237
* P < 0.05 was considered statistically significant. OR = odds ratio; 95% CI = 95% confidence interval, LA = leukoaraiosis.
Table 4 The correlation of other risk factors with LA and their effects on LA when combined with retinopathy (linear regression) Linear regression Variables
Coefficient
95% CI
P value
Age Coronary heart disease Carotid artery plaque C-reactive protein Systolic pressure
0.036 0.388 0.547 0.069 0.013
0.020–0.051 0.041–0.734 0.199–0.895 0.037–0.101 0.004–0.023
<0.0018* 0.029* 0.002* <0.001* 0.006*
* P < 0.05 was considered statistically significant. 95% CI = 95% confidence interval, LA = leukoaraiosis.
Fig. 3. Leukoaraiosis grades of patients with or without various subtypes of retinopathy. n = 34 without retinopathy, n = 99 with A–V nicking, n = 15 with microaneurysm, n = 43 with hard exudate, n = 8 with cotton-wool spot, n = 11 with hemorrhage. (Mean ± SD; one-way ANOVA, Tukey’s test; ⁄⁄p < 0.001 vs. non retinopathy).
In addition, using Pearson’s correlation and linear regression analysis, we found correlations between the severity of LA and age (r = 0.331, p < 0.001), coronary heart disease (r = 0.169, p = 0.029), C-reactive protein (r = 0.169, p = 0.029), carotid artery plaque (r = 0.234, p = 0.002) and systolic blood pressure (r = 0.209, p = 0.006, Table 4). To further expand the predictive value of retinopathy for LA, we combined retinopathy and these risk factors to perform logistic regression analysis. According to the median value of age (cutoff value: 59 years), all eligible patients could be divided into four groups: old age (>59 years) with retinopathy, old age without retinopathy, young age (659 years) with retinopathy and young age without retinopathy. The patients with retinopathy and who were >59 years of age had a 2.870-fold greater risk (95% CI: 1.149–7.171, p = 0.024) of LA than did those who were 659 years of age without retinopathy. Similarly, we found that the retinopathy patients with C-reactive protein >5.8 mmol/L (OR = 23.021) or systolic pressure >139 mmHg (OR = 5.909) were more prone to develop LA. Moreover, the co-occurrence of retinopathy and coronary heart disease (OR = 5.737) or carotid artery plaque (OR = 4.248) significantly increased the risk of LA (Table 5).
4. Discussion In the present study, we provide the first direct evidence that retinopathy is associated with the presence and severity of LA. Using logistic regression analysis, we also found that retinopathy was an independent risk factor for LA, and when combined with age, carotid artery plaque, systolic pressure, C-reactive protein, or coronary heart disease, it can significantly increase the risk of LA. More than a quarter of the population over the age of 65 develops LA, and young adults suffer from this disease as well [19]. Considering the large size of the affected population and the close association of LA with stroke and dementia, it is an important condition to monitor. Because of its common pathological alterations and risk factors, assessing retinopathy may provide a means for monitoring LA routinely and conveniently [13]. However, the premise behind this possibility necessitates finding a solid association between retinopathy and LA. Previous studies have reported that retinal microangiopathy and LA are linked to microvessel disease [20], and the branching angles of the retinal arterioles correlate with cerebral small vessel disease [21], implying a correlation between retinopathy and LA. Our study showed that the incidence of retinopathy was significantly increased in patients with LA, and LA occurred more frequently in patients with any type of retinopathy than in patients without retinopathy. These findings indicate that retinopathy is indeed correlated with LA. Hypertension plays a pivotal role in the pathogenesis of LA [22,23]. In our study, we found that systolic pressure but not diastolic pressure is strongly associated with the severity of LA. Combining high systolic pressure with retinopathy significantly
Please cite this article in press as: Wei W et al. Correlation of retinopathy with leukoaraiosis in patients with anterior circulation infarcts. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2016.02.042
5
W. Wei et al. / Journal of Clinical Neuroscience xxx (2016) xxx–xxx Table 5 The correlation of other risk factors with LA and their effects on LA when combined with retinopathy (Bivariate logistic regression) Bivariate logistic regression Variables
Combined variables
OR
95% CI
P value
Age
Retinopathy and age > 59 years (relative to non-retinopathy and age 6 59 years) Non-retinopathy and age > 59 years (relative to non-retinopathy and age 6 59 years) Retinopathy and age 6 59 (relative to non-retinopathy and age 6 59 years)
2.870 1.034 1.975
1.149–7.171 1.058–8.352 1.242–7.947
0.024* 0.143 0.083
Coronary heart disease
Retinopathy and CHD (relative to non-retinopathy and non-CHD) Non-Retinopathy and CHD (relative to non-retinopathy and non-CHD) Retinopathy and non-CHD (relative to non-retinopathy and non-CHD)
5.737 1.662 2.214
2.207–14.917 1.985–16.321 2.471–15.874
<0.001* 0.154 0.014*
Carotid artery plaque
Retinopathy and CAP (relative to non-retinopathy and non-CAP) Non-retinopathy and CAP (relative to non-retinopathy and non-CAP) Retinopathy and non-CAP (relative to non-retinopathy and non-CAP)
4.248 1.325 1.874
1.868–9.659 1.964–9.848 1.404–10.125
0.001* 0.076 0.055
C-reactive protein
Retinopathy and CRP > 5.8 mmol/L (relative to non- retinopathy and CRP 6 5.8 mmol/L) Non-retinopathy and CRP > 5.8 mmol/L (relative to non- retinopathy and CRP 6 5.8 mmol/L) Retinopathy and CRP 6 5.8 mmol/L (relative to non- retinopathy and CRP 6 5.8 mmol/L)
23.021 1.425 4.421
7.035–75.334 5.021–65.334 3.035–50.332
0.0001* 0.075 0.051
Systolic pressure
Retinopathy and SP > 139 mmol/L (relative to non- retinopathy and SP 6 139 mmol/L) Non-retinopathy and SP > 139 mmol/L (relative to non- retinopathy and SP 6 139 mmol/L) Retinopathy and SP 6 139 mmol/L (relative to non- retinopathy and SP 6 139 mmol/L)
5.909 1.210 2.201
2.399–14.556 2.046–16.453 1.784–17.036
0.0001* 0.089 0.040*
In each bivariate logistic regression analysis, the whole population was divided into four groups based on the combinations of various conditions of retinopathy and another variable. For the continuous variables, the patients were dichotomized according to their median value, and for categorical variables, the patients were grouped according to its presence or absence of the variable. * P < 0.05 was considered statistically significant. CAP = carotid artery plaque, CHD = coronary heart disease, CRP = C-reactive protein, LA = leukoaraiosis, SP = systolic pressure, 95% CI = 95% confidence interval.
increases the risk for LA. This effect may occur because high systolic pressure causes chronic sclerosis of the cerebral arteries and/or arterioles, resulting in a significant decrease in the blood supply to the cerebral white matter region and subsequent demyelination. Considering that hypertension is also a risk factor for retinopathy, the prevention and treatment of hypertension, particularly the control of systolic pressure, is crucial for halting LA development. Our study found that CRP level was significantly elevated in the LA group compared with the non-LA group, and multivariate logistic regression revealed that CRP is an independent risk factor for LA. Surprisingly, compared with patients who present CRP 65.8 mmol/L without retinopathy, the patients with retinopathy and CRP >5.8 mmol/L had an approximately 23-fold greater risk for LA, which is far greater than the individual effect of each variable on LA. Accumulating evidence suggests that inflammation is an essential pathological process underlying microvascular disease, and CRP, as a well-recognized inflammatory factor, is involved in this process [24]. Thus, the effect of retinopathy on LA may be amplified by its interaction with inflammation factors, at least when abnormal CRP is present, or that the impact of CRP on LA is influenced by retinopathy. Therefore, intervention strategies targeting inflammatory factors, particularly CRP, may become an effective treatment for cerebral microvascular disease. Moreover, timely and routine monitoring of CRP and retinopathy may inform evaluation of the development and progression of LA. The association between LA and carotid artery disease remains controversial. Research based on the North American Symptomatic Carotid Endarterectomy Trial found no association between these conditions [25], and a cross-sectional study with a large sample size also confirmed that they are two unrelated disease processes [26]. However, recent studies showed that the presence and severity of leukoaraiosis was closely related to the degree of carotid stenosis and the presence of specific types of plaque (fatty, mixed, and calcified) [27], as well as carotid artery wall thickness [28]. In our study, we found that carotid artery plaque was significantly correlated with LA, and its presence in combination with retinopathy increased the risk of LA although there has no difference in the frequency of carotid artery plaque between the LA and non-LA
groups. The results of the present study are likely due to selection bias of the enrolled participants. Patients with anterior circulation infarcts, rather than those without stroke history, were included. This choice may have inflated the effects of carotid artery disease on LA because of the direct correlation with anterior circulation infarcts. Based on our findings, we suggest that early interventions for the treatment of carotid artery plaque and retinopathy may offer clinical benefits for the prevention of LA and subsequent stroke. Coronary artery disease in combination with retinopathy significantly increased cerebral susceptibility to LA (Table 5). Thus, the prevention and treatment of coronary artery disease not only decreases its direct effect, but also perhaps reduces the risk of LA.
5. Conclusion In conclusion, we found that retinopathy was strongly associated with LA and had the capacity to be an independent risk factor. Considering that retinal photography is an easy, economical and non-invasive method, it will be valuable to the assessment and research of LA and other brain microvascular diseases. The present study is retrospective and includes a relatively small sample size, and bias due to participant selection may exist. A prospective study with a larger sample size is needed to further confirm and elucidate the association between retinopathy and LA.
Conflicts of Interest/Disclosures The authors declare that they have no financial or other conflicts of interest in relation to this research and its publication.
Acknowledgement This research was supported by the National Natural Science Foundation of China [No. 81171100] [No. 31000461], and the Beijing Natural Science Foundation [No. 7153176] [No.7123230].
Please cite this article in press as: Wei W et al. Correlation of retinopathy with leukoaraiosis in patients with anterior circulation infarcts. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2016.02.042
6
W. Wei et al. / Journal of Clinical Neuroscience xxx (2016) xxx–xxx
References [1] Henninger N, Khan MA, Zhang J, et al. Leukoaraiosis predicts cortical infarct volume after distal middle cerebral artery occlusion. Stroke 2014;45:689–95. [2] Li C, Ling X, Liu S, et al. Abnormalities of magnetic resonance spectroscopy and diffusion tensor imaging are correlated with executive dysfunction in patients with ischemic leukoaraiosis. J Clin Neurosci 2012;19:718–22. [3] Costello CA, Campbell BC, Perez de la Ossa N, et al. Age over 80 years is not associated with increased hemorrhagic transformation after stroke thrombolysis. J Clin Neurosci 2012;19:360–3. [4] Smith EE. Leukoaraiosis and stroke. Stroke 2010;41:S139–43. [5] Kim G-M, Park K-Y, Avery R, et al. Extensive leukoaraiosis is associated with high early risk of recurrence after ischemic stroke. Stroke 2014;45:479–85. [6] Arsava E, Rahman R, Rosand J, et al. Severity of leukoaraiosis correlates with clinical outcome after ischemic stroke. Neurology 2009;72:1403–10. [7] Caprio FZ, Maas MB, Rosenberg NF, et al. Leukoaraiosis on magnetic resonance imaging correlates with worse outcomes after spontaneous intracerebral hemorrhage. Stroke 2013;44:642–6. [8] Won YS, Chung P-W, Kim YB, et al. Leukoaraiosis predicts poor outcome after spontaneous supratentorial intracerebral hemorrhage. Eur Neurol 2009;64:253–7. [9] Lamar M, Price CC, Libon DJ, et al. Alterations in working memory as a function of leukoaraiosis in dementia. Neuropsychologia 2007;45:245–54. [10] Gootjes L, Teipel S, Zebuhr Y, et al. Regional distribution of white matter hyperintensities in vascular dementia, Alzheimer’s disease and healthy aging. Dement Geriatr Cogn Disord 2004;18:180–8. [11] Hogervorst E, Ribeiro HM, Molyneux A, et al. Plasma homocysteine levels, cerebrovascular risk factors, and cerebral white matter changes (leukoaraiosis) in patients with Alzheimer disease. Arch Neurol 2002;59:787–93. [12] Wong TY, Klein R, Klein BE, et al. Retinal microvascular abnormalities and their relationship with hypertension, cardiovascular disease, and mortality. Surv Ophthalmol 2001;46:59–80. [13] Kwa VI. Our eyes: windows to our souls or crystal balls? Lancet Neurol 2006;5:108–10. [14] Goto I, Kimoto K, Katsuki S, et al. Pathological studies on the intracerebral and retinal arteries in cerebrovascular and noncerebrovascular diseases. Stroke 1975;6:263–9.
[15] Schmidt R, Petrovic K, Ropele S, et al. Progression of leukoaraiosis and cognition. Stroke 2007;38:2619–25. [16] Enzinger C, Fazekas F, Ropele S, et al. Progression of cerebral white matter lesions – clinical and radiological considerations. J Neurol Sci 2007;257:5–10. [17] Baker ML, Hand PJ, Wang JJ, et al. Retinal signs and stroke: revisiting the link between the eye and brain. Stroke 2008;39:1371–9. [18] Baker ML, Hand PJ, Wang JJ, et al. Retinal signs and stroke revisiting the link between the eye and brain. Stroke 2008;39:1371–9. [19] Putaala J, Kurkinen M, Tarvos V, et al. Silent brain infarcts and leukoaraiosis in young adults with first-ever ischemic stroke. Neurology 2009;72:1823–9. [20] Rodríguez I, Lema I, Blanco M, et al. Research: vascular retinal, neuroimaging and ultrasonographic markers of lacunar infarcts. Int J Stroke 2010;5:360–6. [21] Doubal FN, De Haan R, MacGillivray TJ, et al. Retinal arteriolar geometry is associated with cerebral white matter hyperintensities on magnetic resonance imaging. Int J Stroke 2010;5:434–9. [22] Inzitari D, Giordano GP, Ancona AL, et al. Leukoaraiosis, intracerebral hemorrhage, and arterial hypertension. Stroke 1990;21:1419–23. [23] Folsom AR, Yatsuya H, Mosley TH, et al. Risk of intraparenchymal hemorrhage with magnetic resonance imaging-defined leukoaraiosis and brain infarcts. Ann Neurol 2012;71:552–9. [24] Joshi MS, Tong L, Cook AC, et al. Increased myocardial prevalence of C-reactive protein in human coronary heart disease: direct effects on microvessel density and endothelial cell survival. Cardiovasc Pathol 2012;21:428–35. [25] Streifler JY, Eliasziw M, Benavente OR, et al. Lack of relationship between leukoaraiosis and carotid artery disease. Arch Neurol 1995;52:21–4. [26] Schulz UG, Grüter BE, Briley D, et al. Leukoaraiosis and increased cerebral susceptibility to ischemia: lack of confounding by carotid disease. J Am Heart Assoc 2013;2:e000261. [27] Saba L, Sanfilippo R, Pascalis L, et al. Carotid artery abnormalities and leukoaraiosis in elderly patients: evaluation with MDCT. AJR Am J Roentgenol 2009;192:W63–70. [28] Saba L, Pascalis L, Sanfilippo R, et al. Carotid artery wall thickness and leukoaraiosis: preliminary results using multidetector row CT angiography. Am J Neuroradiol 2011;32:955–61.
Please cite this article in press as: Wei W et al. Correlation of retinopathy with leukoaraiosis in patients with anterior circulation infarcts. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2016.02.042
Contents lists available at ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Clinical Study
Correlation of retinopathy with leukoaraiosis in patients with anterior circulation infarcts Wei Wei a, Zhenxi Xia a, Hongyan Gao b, Jingjing Gong a, Liping Yan c, Yonghua Huang a,⇑, Fei Chen a, Weiwei Zhang a a b c
Department of Neurology, PLA Army General Hospital, No.5 Nan men Cang, Dongcheng District, Beijing 100700, China Department of Medical Affairs, General Hospital of Beijing Command PLA, Dongcheng District, Beijing, China Department of Neurology, TangshanWorker Hospital, Tangshan, China
a r t i c l e
i n f o
Article history: Received 20 November 2015 Accepted 14 February 2016 Available online xxxx Keywords: Anterior circulation infarcts Brain microvascular diseases Leukoaraiosis Retinopathy Risk factors
a b s t r a c t Although retinal and cerebral microvessels share similar embryological, anatomical and physiological characteristics, the correlation between retinopathy and leukoaraiosis (LA), a type of brain microvascular disease, is unclear. In the present study, the sample included 213 patients admitted to the department of neurology from January 2012 through October 2012. MRI and retinal photography were performed within 48 hours of hospitalization, and patient demographics, comorbidities, preadmission medications and laboratory data were collected. MRI images were used to divide the patients into LA and non-LA groups. Using multivariate binary logistic regression, the effects of retinopathy on LA were investigated. Of the 213 patients enrolled, 168 were included in this study (LA, n = 108; non-LA, n = 60). Hypertension, coronary heart disease and carotid artery plaque were more common in the LA group, and these patients showed higher blood levels of C-reactive protein, homocysteine and triglycerides. The incidence of retinopathy was significantly increased in the LA group compared with the non-LA group, and there was a significant correlation between the severity of LA and incidence of retinopathy. Retinopathy is an independent risk factor for LA and can significantly increase the risk of LA when combined with age, coronary heart disease, C-reactive protein, carotid artery plaque or systolic pressure. Taken together, retinopathy is associated with LA in patients with anterior circulation infarcts. Retinopathy is an independent risk factor for LA and an increase the risk of LA, and thus facilitating the evaluation of LA. Ó 2016 Published by Elsevier Ltd.
1. Introduction Leukoaraiosis (LA), defined as a rarefaction or ‘‘araiosis” of the brain white matter, was first proposed by Hachinski in 1987 as a neuroimaging diagnostic term [1]. It is showing up as decreased signal density on CT scan and increased density on T2-wighted MRI/fluid attenuated inversion recovery (FLAIR) MRI sequences performed as MRI scans [2,3]. It mainly appears in the subcortical white matter, centrum semiovale, periventricular and/or deep white matter, where they receive the blood supply primarily from the terminal arteries of the middle cerebral artery and the terminal branches of the choroidal artery. It is well recognized that LA is a type of brain microvascular disease with important clinical implications [4]. Accumulating evidences have demonstrated that the LA commonly occurs in stroke patients and closely associated with the occurrence and clinical outcome of both ischemic [1,5,6] and ⇑ Corresponding author. Tel.: +86 010 84008399; fax: +86 010 64056642. E-mail address: [email protected] (Y. Huang).
hemorrhagic [7,8] stroke. As a composite marker of baseline brain integrity, the presence or severity of LA is also strongly associated with the risk of vascular dementia [4,9] and Alzheimer’s disease [10,11]. Retinal and cerebral microvessels share a common embryological origin, anatomical basis and physiological characteristics [12,13]. A postmortem study of stroke patients have revealed that the pathological alterations of retinal vessels are similar to those of cerebral microvessels [14]. Incomplete white matter infarction caused by stenosis or occlusion of small cerebral vessels with sudden or chronic ischemia is considered to play a crucial role in the pathogenesis of LA [15]. LA progression over time is substantial, especially in patients with certain phenotypes [16]. During LA progression in cerebral microvessels, the stenosis or occlusion may also occur in the retina as there is a close relationship between the microvessels in the eye and the brain. Many retinal conditions, including hypertensive retinopathy, diabetic retinopathy, and retinal arteriolar emboli, have been reported to be associated with stroke, reflecting concomitant pathophysiological processes
http://dx.doi.org/10.1016/j.jocn.2016.02.042 0967-5868/Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Wei W et al. Correlation of retinopathy with leukoaraiosis in patients with anterior circulation infarcts. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2016.02.042
2
W. Wei et al. / Journal of Clinical Neuroscience xxx (2016) xxx–xxx
affecting both the eye and the brain [17]. However, the relationship between retinal conditions and LA has not been studied yet. In this study, a total of 213 anterior circulation infarcts patients were recruited and divided according to the presence of LA. MRI scans and retinal photography were performed within 48 hours of hospitalization, and patient demographics, comorbidities and laboratory data were collected. Image analysis of LA and the assessment of retinopathy were carried out by two experienced neurologists with high inter-rater agreement. By comparing retinopathy and other confounding factors in the LA and non-LA groups and applying multivariate logistic regression, the correlation between retinopathy and LA was investigated. 2. Methods 2.1. Patients In this study, we consecutively recruited 213 patients with anterior circulation infarcts who were admitted to our hospital from January 2012 to October 2012. All patients underwent MRI scans and retinal photography within 48 hours of admission. According to the presence of white matter hyperintensity on the FLAIR sequence and T2-weighted MRI, the subjects were divided into two groups: an LA group and a non-LA group. 2.2. Inclusion criteria The selected patients were in the age of 40–80 years and able to cooperate with the examiner to complete the MRI scan in the supine position or retinal photography in the seated position. 2.3. Exclusion criteria The patients were not eligible if they presented the following: (1) white matter disease caused by multiple sclerosis or infectious, metabolic, toxic, autoimmune and metastatic diseases; (2) history of hemorrhagic stroke, brain tumors or cerebral hemorrhage; (3) concomitant systematic disorder that potentially affects the optic fundus (for example, severe liver, kidney or heart failure; severe infection; malignant disease; systemic lupus erythematosus; or hereditary disease); (4) local eye disorders that could cause an optic fundus disease, such as various eye inflammatory responses or eye surgeries (e.g., cataract extraction or laser surgery) within six months prior to admission; (5) narrow anterior chamber angle; (6) allergy to mydriatic agents, or (7) disturbance of consciousness. 2.4. Collection of clinical data Patient demographics, comorbidities, preadmission medications and laboratory data were collected within 48 hours of hospitalization. Specifically, we assessed the presence of hypertension (at least two blood pressure readings of P140/90 mmHg on separate days measured after a 20-minute rest or taking antihypertensive medication), diabetes mellitus (at least two random serum glucose results >11.1 mmol/L or a fasting serum glucose level >7.0 mmol/L, a specialist’s diagnosis or the use of diabetic medication), coronary heart disease (a specialist’s diagnosis or the use of medication), smoking history (>1 cigarette per day for >1 year), and drinking history (100 g alcohol intake per day for >5 years), and the data were recorded based on the corresponding criteria. Lacunar infarction and atherosclerotic carotid plaque were documented after the MRI scan and ultrasonography, and basic biochemical indexes were collected, including fasting serum glucose, cholesterol, triglycerides, C-reactive protein (CRP) and homocysteine.
2.5. Neuroimaging protocol and image analysis of leukoaraiosis MRI scans were obtained from all patients using a 3.0 T SIEMENS MRI scanner, with sequences that included FLAIR and T1-weighted and T2-weighted gradient echo. The severity of LA was graded based on the periventricular white matter hyperintensity according to the Fazekas scale: 0, absent; 1, caps or pencil-thin lining; 2, smooth halo; or 3, irregular periventricular hyperintensity extending into the deep white matter. LA was reviewed independently by 2 experienced, board-certified neurologists who were blinded to both the research design and the clinical data. To evaluate the degree of inter-rater agreement, a Kappa statistic was calculated from the findings of 30 randomly chosen patients. The Kappa value was >0.80, indicating that the two neurologists had almost perfect consistency when determining the presence or extent of LA. 2.6. Retinal photography and assessment Retinal photography was conducted in all patients after pharmacological mydriasis by experienced ophthalmologists, and fundus color transparencies were taken (50° field centered on the optic disc), followed by an assessment of the presence of retinopathy by 2 trained graders (the neurologists mentioned above). The determination of retinal arteriovenous (AV) nicking, microaneurysms, cotton-wool spots, hard exudates, and retinal hemorrhage was based on standard protocols [18]. For quality control, a random subsample of 30 retinal photographs was examined and the assessment results from the 2 graders were tested using a weighted Kappa analysis. The Kappa values revealed that the inter-grader agreement was also high. 2.7. Standard protocol approval and patient consent The design and implementation of this research was approved and documented by the medical ethics committee at our hospital, and written informed consent was obtained from the patient or a legally authorized representative in all cases. 2.8. Statistics To evaluate the degree of inter-rater agreement, we examined weighted Kappa statistics in a cohort of 30 randomly chosen patients. According to the Kappa values, the inter-rater agreement was interpreted as slight (0.01–0.2), fair (0.21–0.4), moderate (0.41–0.6), substantial (0.61–0.8), or almost perfect (0.81–1.0). Continuous variables are presented as the mean ± standard deviation (SD), and categorical variables are reported as percentages. Continuous variables were assessed for a normal distribution using the one-sample Kolmogorov–Smirnov test. Normally distributed continuous variables were processed using independent-sample t-tests, and non-normally distributed continuous data were compared using the Mann–Whitney U test. Categorical variables were analyzed using the v2 test. To evaluate the correlation between LA and retinopathy, binary logistic regression was used, and an analysis of the risk factors for LA or retinopathy was also performed using multivariate logistic regression. Two-sided tests were used for all analyses, and P < 0.05 was defined as significant. All statistical analyses were performed using IBM SPSS version 18.0. 3. Results A total of 213 patients with anterior circulation infarcts were enrolled in this study. Among these patients, eight were excluded due to severe systematic diseases (three for severe heart failure,
Please cite this article in press as: Wei W et al. Correlation of retinopathy with leukoaraiosis in patients with anterior circulation infarcts. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2016.02.042
3
W. Wei et al. / Journal of Clinical Neuroscience xxx (2016) xxx–xxx
four for malignant disease, and one for systemic erythematosus), and 13 were excluded due to various levels of disturbed consciousness (n = 12) or disturbed consciousness in combination with posterior circulation infarcts (n = 1). Of the remaining 24 patients who were excluded, 14 did not undergo brain MRI within the first 24 hours of admission; three declined to participate in this study; and seven did not complete the retinal photography. The remaining 168 patients formed the final eligible population (n = 108 in the LA group and n = 60 in the non-LA group). The demographic and clinical characteristics of the LA and non-LA groups are shown in Table 1 and Figure 1. There were no differences in the sex ratio, smoking or alcohol history, nor in the incidence of diabetes between the LA and non-LA groups. Hypertension and carotid artery plaque were more frequently observed in the LA group compared with the non-LA group. The patients with LA had higher levels of C-reactive protein (P < 0.01), homocysteine (P < 0.01) and triglycerides (P < 0.04) in the blood and were more likely to have high blood pressure, both systolic (P < 0.01) and diastolic (P < 0.04) (Fig. 1). No significant differences were observed in fasting serum glucose, high-density lipoprotein, or low-density lipoprotein between the two groups.
The incidence of retinopathy was significantly increased in the LA group compared with the non-LA group (88.0% vs. 65.0%, p < 0.001). With the exception of cotton-wool spot (3.7% vs. 6.7%), most signs, including A-V nicking (70.4% vs. 38.3%), microaneurysm (11.1% vs. 5.0%), hard exudate (27.6% vs. 15.0%) and hemorrhage (7.4% vs. 5.0%), were more frequently observed in the LA group than in the non-LA group (Table 2). Among these signs, differences in the incidences of A-V nicking and hard exudate were significant (Table 2). Moreover, LA occurred more frequently in patients with any type of retinopathy than in patients without retinopathy (Fig. 2). Patients with any retinopathies, particularly microaneurysm (p = 0.002) and hard exudate (p < 0.001), exhibited more severe LA than patients without retinopathy (Fig. 3). There was a significant correlation between the severity of LA and the incidence of retinopathy when all 168 patients were analyzed together (r = 0.186, p = 0.016). The patients with retinopathy had a 3.935fold greater risk (95% confidence interval [CI]: 1.793–8.633, p < 0.001) of LA than those without retinopathy. After adjustment for the other risk factors of LA, the odds ratio still remained at 3.745 (95% CI: 1.100–12.752, p = 0.040, Table 3), indicating that retinopathy is an independent risk factor for LA.
Table 1 Demographic characteristics of leukoaraiosis and non-leukoaraiosis groups Variables Sex
Male, n (%) Female, n (%)
Age (years), mean ± SD Smoking history, n (%) Alcohol history, n (%) Hypertension, n (%) Diabetes mellitus, n (%) Coronary heart disease, n (%) Lacunar infarction, n (%) Carotid artery plaque, n (%)
LA (n = 108)
Non-LA (n = 60)
P value
53 (49.1) 55 (50.9) 65.55 ± 11.16 41 (38.0) 30 (27.8) 75 (68.8) 50 (46.5) 48 (44.3) 70 (64.8) 68 (57.7)
29 (48.3) 31 (51.7) 62.73 ± 9.37 22 (36.7) 17 (28.3) 32 (53.3) 18 (29.0) 19 (31.7) 29 (48.3) 30 (50.0)
0.88 0.71 0.10 0.86 0.67 0.04* 0.22 0.11 0.04* 0.10
* P < 0.05 was considered statistically significant. LA = leukoaraiosis.
Fig. 1. The clinical characteristics of patients in the leukoaraiosis and non-leukoaraiosis groups.
Table 2 Incidence of each type of retinopathy, n (%) Group
N
A-V nicking
Microaneurysm
Hard exudate
Cotton-wool spot
Hemorrhage
LA Non-LA P value
108 60
76 (70.4) 23 (38.3) 0.001*
12 (11.1) 9 (5.4) 0.18
34 (27.6) 9 (15.4) 0.019*
4 (3.7) 4 (8.0) 0.39
8 (7.4) 3 (5.0) 0.55
* P < 0.05 was considered statistically significant. LA = leukoaraiosis.
Please cite this article in press as: Wei W et al. Correlation of retinopathy with leukoaraiosis in patients with anterior circulation infarcts. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2016.02.042
4
W. Wei et al. / Journal of Clinical Neuroscience xxx (2016) xxx–xxx Table 3 Effects of retinopathy and other risk factors on LA (logistic regression) P value
Fig. 2. Percentage of patients with various leukoaraiosis grades for each subtype of retinopathy. All patients in the leukoaraiosis and non-leukoaraiosis groups were analyzed together.
Retinopathy Age Smoking history Alcohol history Hypertension Diabetes mellitus Coronary heart disease Carotid artery plaque Systolic pressure Diastolic pressure Fasting serum glucose Triglyceride Cholesterol High-density lipoprotein Low-density lipoprotein C-reactive protein Homocysteine
*
0.035 0.042* 0.297 0.651 0.316 0.849 0.564 0.983 0.499 0.161 0.466 0.287 0.704 0.931 0.674 <0.001* 0.105
OR
95% CI
3.745 2.843 1.990 0.752 0.605 1.107 1.318 0.989 1.012 1.039 0.896 1.451 0.793 1.093 1.298 1.391 2.290
1.100–12.752 0.993–1.094 0.546–7.254 0.219–2.586 0.227–1.615 0.387–3.173 0.515–3.371 0.351–2.788 0.977–1.048 0.985–1.097 0.668–1.203 0.732–2.878 0.240–2.619 0.145–8.262 0.385–4.376 1.213–1.595 0.841–6.237
* P < 0.05 was considered statistically significant. OR = odds ratio; 95% CI = 95% confidence interval, LA = leukoaraiosis.
Table 4 The correlation of other risk factors with LA and their effects on LA when combined with retinopathy (linear regression) Linear regression Variables
Coefficient
95% CI
P value
Age Coronary heart disease Carotid artery plaque C-reactive protein Systolic pressure
0.036 0.388 0.547 0.069 0.013
0.020–0.051 0.041–0.734 0.199–0.895 0.037–0.101 0.004–0.023
<0.0018* 0.029* 0.002* <0.001* 0.006*
* P < 0.05 was considered statistically significant. 95% CI = 95% confidence interval, LA = leukoaraiosis.
Fig. 3. Leukoaraiosis grades of patients with or without various subtypes of retinopathy. n = 34 without retinopathy, n = 99 with A–V nicking, n = 15 with microaneurysm, n = 43 with hard exudate, n = 8 with cotton-wool spot, n = 11 with hemorrhage. (Mean ± SD; one-way ANOVA, Tukey’s test; ⁄⁄p < 0.001 vs. non retinopathy).
In addition, using Pearson’s correlation and linear regression analysis, we found correlations between the severity of LA and age (r = 0.331, p < 0.001), coronary heart disease (r = 0.169, p = 0.029), C-reactive protein (r = 0.169, p = 0.029), carotid artery plaque (r = 0.234, p = 0.002) and systolic blood pressure (r = 0.209, p = 0.006, Table 4). To further expand the predictive value of retinopathy for LA, we combined retinopathy and these risk factors to perform logistic regression analysis. According to the median value of age (cutoff value: 59 years), all eligible patients could be divided into four groups: old age (>59 years) with retinopathy, old age without retinopathy, young age (659 years) with retinopathy and young age without retinopathy. The patients with retinopathy and who were >59 years of age had a 2.870-fold greater risk (95% CI: 1.149–7.171, p = 0.024) of LA than did those who were 659 years of age without retinopathy. Similarly, we found that the retinopathy patients with C-reactive protein >5.8 mmol/L (OR = 23.021) or systolic pressure >139 mmHg (OR = 5.909) were more prone to develop LA. Moreover, the co-occurrence of retinopathy and coronary heart disease (OR = 5.737) or carotid artery plaque (OR = 4.248) significantly increased the risk of LA (Table 5).
4. Discussion In the present study, we provide the first direct evidence that retinopathy is associated with the presence and severity of LA. Using logistic regression analysis, we also found that retinopathy was an independent risk factor for LA, and when combined with age, carotid artery plaque, systolic pressure, C-reactive protein, or coronary heart disease, it can significantly increase the risk of LA. More than a quarter of the population over the age of 65 develops LA, and young adults suffer from this disease as well [19]. Considering the large size of the affected population and the close association of LA with stroke and dementia, it is an important condition to monitor. Because of its common pathological alterations and risk factors, assessing retinopathy may provide a means for monitoring LA routinely and conveniently [13]. However, the premise behind this possibility necessitates finding a solid association between retinopathy and LA. Previous studies have reported that retinal microangiopathy and LA are linked to microvessel disease [20], and the branching angles of the retinal arterioles correlate with cerebral small vessel disease [21], implying a correlation between retinopathy and LA. Our study showed that the incidence of retinopathy was significantly increased in patients with LA, and LA occurred more frequently in patients with any type of retinopathy than in patients without retinopathy. These findings indicate that retinopathy is indeed correlated with LA. Hypertension plays a pivotal role in the pathogenesis of LA [22,23]. In our study, we found that systolic pressure but not diastolic pressure is strongly associated with the severity of LA. Combining high systolic pressure with retinopathy significantly
Please cite this article in press as: Wei W et al. Correlation of retinopathy with leukoaraiosis in patients with anterior circulation infarcts. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2016.02.042
5
W. Wei et al. / Journal of Clinical Neuroscience xxx (2016) xxx–xxx Table 5 The correlation of other risk factors with LA and their effects on LA when combined with retinopathy (Bivariate logistic regression) Bivariate logistic regression Variables
Combined variables
OR
95% CI
P value
Age
Retinopathy and age > 59 years (relative to non-retinopathy and age 6 59 years) Non-retinopathy and age > 59 years (relative to non-retinopathy and age 6 59 years) Retinopathy and age 6 59 (relative to non-retinopathy and age 6 59 years)
2.870 1.034 1.975
1.149–7.171 1.058–8.352 1.242–7.947
0.024* 0.143 0.083
Coronary heart disease
Retinopathy and CHD (relative to non-retinopathy and non-CHD) Non-Retinopathy and CHD (relative to non-retinopathy and non-CHD) Retinopathy and non-CHD (relative to non-retinopathy and non-CHD)
5.737 1.662 2.214
2.207–14.917 1.985–16.321 2.471–15.874
<0.001* 0.154 0.014*
Carotid artery plaque
Retinopathy and CAP (relative to non-retinopathy and non-CAP) Non-retinopathy and CAP (relative to non-retinopathy and non-CAP) Retinopathy and non-CAP (relative to non-retinopathy and non-CAP)
4.248 1.325 1.874
1.868–9.659 1.964–9.848 1.404–10.125
0.001* 0.076 0.055
C-reactive protein
Retinopathy and CRP > 5.8 mmol/L (relative to non- retinopathy and CRP 6 5.8 mmol/L) Non-retinopathy and CRP > 5.8 mmol/L (relative to non- retinopathy and CRP 6 5.8 mmol/L) Retinopathy and CRP 6 5.8 mmol/L (relative to non- retinopathy and CRP 6 5.8 mmol/L)
23.021 1.425 4.421
7.035–75.334 5.021–65.334 3.035–50.332
0.0001* 0.075 0.051
Systolic pressure
Retinopathy and SP > 139 mmol/L (relative to non- retinopathy and SP 6 139 mmol/L) Non-retinopathy and SP > 139 mmol/L (relative to non- retinopathy and SP 6 139 mmol/L) Retinopathy and SP 6 139 mmol/L (relative to non- retinopathy and SP 6 139 mmol/L)
5.909 1.210 2.201
2.399–14.556 2.046–16.453 1.784–17.036
0.0001* 0.089 0.040*
In each bivariate logistic regression analysis, the whole population was divided into four groups based on the combinations of various conditions of retinopathy and another variable. For the continuous variables, the patients were dichotomized according to their median value, and for categorical variables, the patients were grouped according to its presence or absence of the variable. * P < 0.05 was considered statistically significant. CAP = carotid artery plaque, CHD = coronary heart disease, CRP = C-reactive protein, LA = leukoaraiosis, SP = systolic pressure, 95% CI = 95% confidence interval.
increases the risk for LA. This effect may occur because high systolic pressure causes chronic sclerosis of the cerebral arteries and/or arterioles, resulting in a significant decrease in the blood supply to the cerebral white matter region and subsequent demyelination. Considering that hypertension is also a risk factor for retinopathy, the prevention and treatment of hypertension, particularly the control of systolic pressure, is crucial for halting LA development. Our study found that CRP level was significantly elevated in the LA group compared with the non-LA group, and multivariate logistic regression revealed that CRP is an independent risk factor for LA. Surprisingly, compared with patients who present CRP 65.8 mmol/L without retinopathy, the patients with retinopathy and CRP >5.8 mmol/L had an approximately 23-fold greater risk for LA, which is far greater than the individual effect of each variable on LA. Accumulating evidence suggests that inflammation is an essential pathological process underlying microvascular disease, and CRP, as a well-recognized inflammatory factor, is involved in this process [24]. Thus, the effect of retinopathy on LA may be amplified by its interaction with inflammation factors, at least when abnormal CRP is present, or that the impact of CRP on LA is influenced by retinopathy. Therefore, intervention strategies targeting inflammatory factors, particularly CRP, may become an effective treatment for cerebral microvascular disease. Moreover, timely and routine monitoring of CRP and retinopathy may inform evaluation of the development and progression of LA. The association between LA and carotid artery disease remains controversial. Research based on the North American Symptomatic Carotid Endarterectomy Trial found no association between these conditions [25], and a cross-sectional study with a large sample size also confirmed that they are two unrelated disease processes [26]. However, recent studies showed that the presence and severity of leukoaraiosis was closely related to the degree of carotid stenosis and the presence of specific types of plaque (fatty, mixed, and calcified) [27], as well as carotid artery wall thickness [28]. In our study, we found that carotid artery plaque was significantly correlated with LA, and its presence in combination with retinopathy increased the risk of LA although there has no difference in the frequency of carotid artery plaque between the LA and non-LA
groups. The results of the present study are likely due to selection bias of the enrolled participants. Patients with anterior circulation infarcts, rather than those without stroke history, were included. This choice may have inflated the effects of carotid artery disease on LA because of the direct correlation with anterior circulation infarcts. Based on our findings, we suggest that early interventions for the treatment of carotid artery plaque and retinopathy may offer clinical benefits for the prevention of LA and subsequent stroke. Coronary artery disease in combination with retinopathy significantly increased cerebral susceptibility to LA (Table 5). Thus, the prevention and treatment of coronary artery disease not only decreases its direct effect, but also perhaps reduces the risk of LA.
5. Conclusion In conclusion, we found that retinopathy was strongly associated with LA and had the capacity to be an independent risk factor. Considering that retinal photography is an easy, economical and non-invasive method, it will be valuable to the assessment and research of LA and other brain microvascular diseases. The present study is retrospective and includes a relatively small sample size, and bias due to participant selection may exist. A prospective study with a larger sample size is needed to further confirm and elucidate the association between retinopathy and LA.
Conflicts of Interest/Disclosures The authors declare that they have no financial or other conflicts of interest in relation to this research and its publication.
Acknowledgement This research was supported by the National Natural Science Foundation of China [No. 81171100] [No. 31000461], and the Beijing Natural Science Foundation [No. 7153176] [No.7123230].
Please cite this article in press as: Wei W et al. Correlation of retinopathy with leukoaraiosis in patients with anterior circulation infarcts. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2016.02.042
6
W. Wei et al. / Journal of Clinical Neuroscience xxx (2016) xxx–xxx
References [1] Henninger N, Khan MA, Zhang J, et al. Leukoaraiosis predicts cortical infarct volume after distal middle cerebral artery occlusion. Stroke 2014;45:689–95. [2] Li C, Ling X, Liu S, et al. Abnormalities of magnetic resonance spectroscopy and diffusion tensor imaging are correlated with executive dysfunction in patients with ischemic leukoaraiosis. J Clin Neurosci 2012;19:718–22. [3] Costello CA, Campbell BC, Perez de la Ossa N, et al. Age over 80 years is not associated with increased hemorrhagic transformation after stroke thrombolysis. J Clin Neurosci 2012;19:360–3. [4] Smith EE. Leukoaraiosis and stroke. Stroke 2010;41:S139–43. [5] Kim G-M, Park K-Y, Avery R, et al. Extensive leukoaraiosis is associated with high early risk of recurrence after ischemic stroke. Stroke 2014;45:479–85. [6] Arsava E, Rahman R, Rosand J, et al. Severity of leukoaraiosis correlates with clinical outcome after ischemic stroke. Neurology 2009;72:1403–10. [7] Caprio FZ, Maas MB, Rosenberg NF, et al. Leukoaraiosis on magnetic resonance imaging correlates with worse outcomes after spontaneous intracerebral hemorrhage. Stroke 2013;44:642–6. [8] Won YS, Chung P-W, Kim YB, et al. Leukoaraiosis predicts poor outcome after spontaneous supratentorial intracerebral hemorrhage. Eur Neurol 2009;64:253–7. [9] Lamar M, Price CC, Libon DJ, et al. Alterations in working memory as a function of leukoaraiosis in dementia. Neuropsychologia 2007;45:245–54. [10] Gootjes L, Teipel S, Zebuhr Y, et al. Regional distribution of white matter hyperintensities in vascular dementia, Alzheimer’s disease and healthy aging. Dement Geriatr Cogn Disord 2004;18:180–8. [11] Hogervorst E, Ribeiro HM, Molyneux A, et al. Plasma homocysteine levels, cerebrovascular risk factors, and cerebral white matter changes (leukoaraiosis) in patients with Alzheimer disease. Arch Neurol 2002;59:787–93. [12] Wong TY, Klein R, Klein BE, et al. Retinal microvascular abnormalities and their relationship with hypertension, cardiovascular disease, and mortality. Surv Ophthalmol 2001;46:59–80. [13] Kwa VI. Our eyes: windows to our souls or crystal balls? Lancet Neurol 2006;5:108–10. [14] Goto I, Kimoto K, Katsuki S, et al. Pathological studies on the intracerebral and retinal arteries in cerebrovascular and noncerebrovascular diseases. Stroke 1975;6:263–9.
[15] Schmidt R, Petrovic K, Ropele S, et al. Progression of leukoaraiosis and cognition. Stroke 2007;38:2619–25. [16] Enzinger C, Fazekas F, Ropele S, et al. Progression of cerebral white matter lesions – clinical and radiological considerations. J Neurol Sci 2007;257:5–10. [17] Baker ML, Hand PJ, Wang JJ, et al. Retinal signs and stroke: revisiting the link between the eye and brain. Stroke 2008;39:1371–9. [18] Baker ML, Hand PJ, Wang JJ, et al. Retinal signs and stroke revisiting the link between the eye and brain. Stroke 2008;39:1371–9. [19] Putaala J, Kurkinen M, Tarvos V, et al. Silent brain infarcts and leukoaraiosis in young adults with first-ever ischemic stroke. Neurology 2009;72:1823–9. [20] Rodríguez I, Lema I, Blanco M, et al. Research: vascular retinal, neuroimaging and ultrasonographic markers of lacunar infarcts. Int J Stroke 2010;5:360–6. [21] Doubal FN, De Haan R, MacGillivray TJ, et al. Retinal arteriolar geometry is associated with cerebral white matter hyperintensities on magnetic resonance imaging. Int J Stroke 2010;5:434–9. [22] Inzitari D, Giordano GP, Ancona AL, et al. Leukoaraiosis, intracerebral hemorrhage, and arterial hypertension. Stroke 1990;21:1419–23. [23] Folsom AR, Yatsuya H, Mosley TH, et al. Risk of intraparenchymal hemorrhage with magnetic resonance imaging-defined leukoaraiosis and brain infarcts. Ann Neurol 2012;71:552–9. [24] Joshi MS, Tong L, Cook AC, et al. Increased myocardial prevalence of C-reactive protein in human coronary heart disease: direct effects on microvessel density and endothelial cell survival. Cardiovasc Pathol 2012;21:428–35. [25] Streifler JY, Eliasziw M, Benavente OR, et al. Lack of relationship between leukoaraiosis and carotid artery disease. Arch Neurol 1995;52:21–4. [26] Schulz UG, Grüter BE, Briley D, et al. Leukoaraiosis and increased cerebral susceptibility to ischemia: lack of confounding by carotid disease. J Am Heart Assoc 2013;2:e000261. [27] Saba L, Sanfilippo R, Pascalis L, et al. Carotid artery abnormalities and leukoaraiosis in elderly patients: evaluation with MDCT. AJR Am J Roentgenol 2009;192:W63–70. [28] Saba L, Pascalis L, Sanfilippo R, et al. Carotid artery wall thickness and leukoaraiosis: preliminary results using multidetector row CT angiography. Am J Neuroradiol 2011;32:955–61.
Please cite this article in press as: Wei W et al. Correlation of retinopathy with leukoaraiosis in patients with anterior circulation infarcts. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2016.02.042