International Journal of Neuroscience, 2015; 125(3): 186–190 Copyright © 2014 Informa Healthcare USA, Inc. ISSN: 0020-7454 print / 1543-5245 online DOI: 10.3109/00207454.2014.951042
ORIGINAL ARTICLE
Higher apolipoprotein B levels are associated with earlier onset of first-ever atherosclerotic stroke Fang Ye, Jing Liu, Shu Yang, and Fu-Qiang Guo Department of Neurology, Sichuan Provincial Academy of Medical Sciences & Sichuan Provincial People’s Hospital, Chengdu, China Background: Many studies have revealed apolipoproteins are risk factors for ischemic stroke, but the influence of apolipoproteins on onset age of first-ever atherosclerotic stroke has not been well investigated. Methods: We recruited 357 qualified participants from consecutive patients with acute ischemic stroke who came to the stroke registry center in Sichuan Provincial People’s Hospital, Chengdu, China. Patients were stratified into tertiles according to the distributions of apoB levels for large artery atherosclerosis (LAA) and small artery atherosclerosis (SAA) groups. The onset age of stroke was analyzed tripartitely in terms of early-onset group, the middlingonset group and the late-onset group. Multinomial logistical regression was used to analyze the associations between the two. Results: The risk of early-onset stroke increased monotonically with higher apoB levels (the second tertile, adjusted OR = 2.61, 95% CI 1.18–5.79 (p = 0.018); the third tertile, adjusted OR = 19.52, 95% CI 5.93–64.31 (p < 0.001)), and patients with the highest tertile of apoB levels had a 9.20 times (95% CI, 2.97–28.53, p < 0.001) increased risk of middling-onset stroke in reference to late onset of stroke. Conclusions: The present study suggests the higher the apolipoprotein B levels are, the earlier an atherosclerotic stroke might occur in a Chinese population. KEYWORDS: stroke, atherosclerotic, onset age, apolipoprotein B, apolipoprotein A1, apolipoprotein B/A1 ratio
Introduction Stroke is a leading cause of mortality and disability worldwide [1]. One third of strokes occur in those below the age of 65 and cause huge burden to the family and society [2]. Study of Brown et al. shows that loss of earnings is multiplied if the stroke occurs earlier in life [3]. Another study reveals that patients with early-onset stroke account for 75% of the total national lifetime costs of stroke but only 30% of the total incidence [4]. Therefore, to explore influencing factors for early-onset stroke is of great importance on which limited studies has focused. Apolipoproteins are consider to be promising predictors for stroke [5–8]. The INTERSTROKE study, a multicenter case–control study of 2337 patients,
Received 2 June 2014; revised 30 July 2014; accepted 30 July 2014 Correspondence: Fang Ye and Fu-Qiang Guo, Department of Neurology, Sichuan Provincial Academy of Medical Sciences & Sichuan, Provincial People’s Hospital, Chengdu 610072, China. Tel: +86 28 87394782/86 28 87394093. Fax: 86 28 87394093. E-mail: [email protected]; [email protected]
186
reveals that apoB/A1 ratio is an independent risk factor of ischemic stroke and the population- attributable risk is 35.2% [5], but few studies have investigated the impact of apolipoproteins (apo) on onset age of ischemic stroke. As apolipoproteins mainly function during the process of atherosclerosis [9], the current study targeted population with first-ever atherosclerotic stroke and was designed to investigate the association between onset age of stroke and serum apolipoproteins adjusting for potential confounders.
Methods and material Patients and relevant examinations and blood tests From 1 March 2013 to 31 January 2014, we prospectively recruited consecutive patients with acute firstever atherosclerotic stroke who came to the stroke registry center in Sichuan Provincial People’s Hospital, Chengdu, China. Informed consent was obtained from all participants, and the study protocol was approved by the Institutional Ethics Committee. The diagnosis
Apolipoproteins and onset age of stroke
was made by experienced neurologists (F. Ye, J. Liu, F.Q. Guo). The inclusion criteria of patients were acute first-ever atherosclerotic stroke with focal neurological impairments and imaging of an acute clinically relevant brain lesion excluding the following: those who had strokes caused by embolism or other undetermined etiologies; those who had suffered from ischemic cerebral stroke or its equivalents such as coronary artery disease and peripheral arterial disease (identified according to clear past history supported by symptoms, signs, treatments, reliable test reports and/or medical files) before the present event; those who had been on statin or fibrate regularly before admission since these drugs affect serum apolipoprotein levels [10]; those who underwent incomplete vascular imaging and laboratory tests and those who did not provide informed consent. Clinical data were collected by trained medical staff (S. Yang) blinded to the classification of the patients. Mean blood pressure was measured by 24 h ambulatory blood pressure monitoring (ABPM) or calculated from at least 10 measures of random BP during hospitalization. Electrocardiogram and echocardiography were performed in all patients to help exclude cardiac embolism; 24 h Holter monitoring was performed when the results of electrocardiogram were abnormal or the presence of arrhythmia was suspected in order to further exclude paroxysmal atrial fibrillation. Fasting concentrations of lipid and apolipoprotein profile were assayed within 48 h from admission by enzymatic techniques (Hitachi 7060 Automatic Biochemical Analyzer, Tokyo, Japan) and immunonephelometry (Dade Behring BN ProSpec, Deerfield, MA), respectively.
Patients’ classification and definition of risk factors All patients took the examinations of non-invasive vascular imaging for both intracranial and extracranial vessels and were classified into the large artery atherosclerosis (LAA) group or the small artery atherosclerosis (SAA) group based on whether there was at least one artery with ≤50% stenosis in the intracranial or extracranial arteries. Vascular imaging findings were reviewed by two independent reviewers (F. Ye and J. Liu) who were blinded to the clinical information. Participants who were taking antihypertensive medication or newly diagnosed hypertension were defined as currently having hypertension. Individuals were classified as having diabetes mullites (DM) if individuals were taking hypoglycemic therapies or newly diagnosed DM. Smoking was defined as currently smoking at least 1 cigarette per day for at least 3 months or at least 100 cigarettes in their lifetime [11]. C
2014 Informa Healthcare USA, Inc.
187
Statistical analysis Data are summarized as mean ± SD or number of subjects (percentage). Differences between two groups were tested using t-test. The multinomial logistical regression was used to analyze associations between serum apolipoprotein levels and onset age of stroke, adjusting for other potential confounders. Results are given by odd ratios (ORs) and its 95% confidential intervals (CIs). A p value 0.05 was considered statistically significant. All analyses were conducted using SPSS Version 19.0.
Results Three hundred fifty-seven patients [mean age at stroke onset, 65.96 ± 11.96 years; male, 229 (64.14%); patients with hypertension, 310 (86.83%); DM, 185 (51.82%); LAA, 163 (45.66%); smoking, 154 (43.14%)] were enrolled from 669 consecutive patients with acute ischemic stroke. Three hundred twelve patients were excluded, among whom, 72 had prior statin or fibrate use; 54 had ischemic stroke or its equivalents before the present event; 145 had strokes caused by cardiac embolism or other less common etiology; 31 had incomplete data and 10 had no informed consent. When analyzing the onset age, three groups were divided according to tertiles of the onset age, defined as the early-onset group with onset age 73 years, respectively. Multinomial logistic regression analyses were performed and the results were showed in Table 1. Among the covariants evaluated, apoB levels as a continuous variable were the only one with significant results when comparing either early-onset or middlingonset group to late-onset one with adjusted ORs of 7.42 (95% CIs 1.70–32.38, p = 0.008) and 7.45 (95% CIs 1.45–31.76, p = 0.007), respectively. Then, the patients were stratified into tertiles according to the distributions of apoB levels or apoB/A1 ratios, respectively, for LAA and SAA groups, because there was obvious difference between the two groups in terms of apoB levels (1.22 ± 0.49 vs. 0.97 ± 0.38, p < 0.001) and apoB/A1 ratios (0.88 ± 0.39 vs. 0.68 ± 0.28, p < 0.001). Multinomial logistic regression analyses were performed among tertiles of the apolipoproteins and onset age of stroke. Table 2 shows the results. When comparing early-onset group to late-onset one, apoB levels were significantly associated with higher risk of early-onset stroke with a dose–response relationship: the second tertile, adjusted OR = 2.61, 95% CIs 1.18–5.79 (p = 0.018); the third tertile, adjusted OR = 19.52, 95% CIs 5.93–64.31 (p < 0.001), when referenced to the first
188
F. Ye et al.
Table 1.
Multinomial logistic regression analyses of apoB, apoA1, apoB/A1 ratio and tertiles of onset age. Early-onset groupa Crude OR (95%CI)
p
Adjusted OR (95%CI)
p
2.74(1.56–4.80) 2.37(1.40–4.00) 1.07(0.51–2.24) 0.97(0.58–1.61) 1.18(0.94–1.49) 1.93(1.33–2.81) 1.31(0.96–1.79) 0.23(0.10–0.51) 2.41(1.28–4.53) 0.20(0.07–0.54) 5.20(2.22–12.18)
0.000 0.001 0.851 0.897 0.151 0.001 0.092 0.000 0.006 0.002 0.000
1.65(0.78–3.48) 1.67(0.85–3.29) N/A 0.82(0.47–1.42) N/A 1.74(1.15–2.62) 0.55(0.26–1.15) 0.45(0.14–1.48) 7.42(1.70–32.38) 0.70(0.16–2.96) 5.66(1.21–26.50)
0.189 0.137
1.05(0.52–2.10) 1.31(0.66–2.61) N/A 1.60(0.93–2.74) N/A 1.20(0.78–1.84) 0.55(0.26–1.15) 0.27(0.08–0.87) 7.45(1.75–31.76) 5.00(1.17–21.30) 2.25(0.48–10.51)
0.901 0.436
Male Smoking Hypertension DM TC, 1-mmol/L difference TG, 1-mmol/L difference LDL-C, 1-mmol/L difference HDL-C, 1-mmol/L difference ApoB, 1-g/L difference ApoA1, 1-g/L difference ApoB/A1ratio, 1-unit differenceb
0.478 0.008 0.112 0.189 0.008 0.623 0.028
Middling-onset groupa Male Smoking Hypertension DM TC, 1-mmol/L difference TG, 1-mmol/L difference LDL-C, 1-mmol/L difference HDL-C, 1-mmol/L difference ApoB, 1-g/L difference ApoA1, 1-g/L difference ApoB/A1 ratio, 1-unit differenceb
1.19(0.71–1.99) 1.24(0.73–2.10) 1.25(0.59–2.67) 1.85(1.10–3.10) 1.17(0.93–1.47) 1.51(1.03–2.21) 1.38(1.01–1.88) 0.61(0.30–1.26) 2.56(1.37–4.83) 1.37(0.54–3.49) 2.87(1.22–5.74)
0.512 0.421 0.564 0.020 0.190 0.034 0.044 0.182 0.003 0.512 0.016
0.088 0.407 0.113 0.028 0.007 0.030 0.304
Bold and italic values indicate those with a significant p value. Apo indicates apolipoprotein; LAA, large artery atherosclerosis; SAA, small artery atherosclerosis; OR, odds ratio; CI, confidence interval; DM, Diabetes mellitus; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; TG, triglyceride; TC, cholesterol. a Compared with late-onset group. b ApoB/A1 ratio and serum levels of apoB and apoA1 entered the multivariate analyses separately.
dicated 1.26 times of that (95% CIs, 0.59–2.70, p = 0.550). For either the second or third tertile of apoB levels, the risk of early-onset stroke was higher than that of middle-onset stroke. As for apoB/A1 ratio, there was significant association between the third tertile of apoB/A1
tertile. When comparing middling-onset group to lateonset one, the dose–response relationship still existed as the third tertile of apoB levels indicated 9.20 times (95% CIs, 2.97–28.53, p < 0.001) increased risk of middlingonset stroke, and the second tertile of apoB levels in-
Table 2.
Logistic regression analyses of tertiles of apoB and apoB/A1 ratio and that of onset age (total 357 patients). Middling and late-onset groups as referent group
Late-onset group as referent group Early-onset group a
Adjusted OR (95% CIs)
Middling-onset group p
1st 2nd 3rd
Referent 1.44 (0.68–3.04) 4.26 (1.55–11.71)
0.346 0.005
1st 2nd 3rd
Referent 2.61 (1.18–5.79) 19.52 (5.93–64.31)
0.018 0.000
a
Adjusted OR (95% CIs) Tertiles of apoB/A1 ratiob Referent 0.96 (0.48–1.92) 2.33 (0.90–6.00) Tertiles of apoB levelsc Referent 1.26 (0.59–2.70) 9.20 (2.97–28.53)
Early-onset group p
Adjusted ORa (95% CIs)
p
0.902 0.081
Referent 1.48 (0.75–2.90) 2.72 (1.17–6.32)
0.254 0.020
0.550 0.000
Referent 2.20 (1.09–4.44) 5.02 (1.92–13.12)
0.028 0.001
Bold and italic values indicate those with a significant p value. a Adjusted for confounders listed in Table 1. b Ranges of apoB/A1 ratio tertile for LAA and SAA group were 0.98 and 0.77, respectively. c Ranges of tertiles of apoB levels for LAA and SAA group were 1.38 g/L and 1.08 g/L, respectively.
International Journal of Neuroscience
Apolipoproteins and onset age of stroke
ratio and increased risk of early-onset stroke (adjusted OR = 4.26, 95% CIs 1.55–11.71, p = 0.005) while negative results were found in the other three comparisons. When comparing early-onset group with middle- and late-onset groups combined, dose–response relationship was revealed for apoB levels with significance (the second tertile, adjusted OR = 2.20, 95% CIs 1.09–4.44, p = 0.028; the third tertile, adjusted OR = 5.02, 95% CIs 1.92–13.12, p = 0.001, in reference to the first tertile) but not for apoB/A1 ratios.
Discussion Our study demonstrated that the risk of early-onset stroke increased monotonically with higher apoB levels (approximately 3 times for the second tertile and 20 times for the third one), and patients with the highest tertile of the apoB levels had a nine-fold increased risk of middling-onset stroke in reference to late onset of stroke. For either the second or third tertile of apoB levels, the risk of early-onset stroke were higher than that of middle-onset stroke. These results indicated that the magnitude of the associations between tertiles of apoB levels and that of onset age was not only dose-responsive, but also age-related (inversely). Thus, a conclusion might be drawn that higher apoB levels might be associated with earlier onset of first-ever acute atherosclerotic stroke in a Chinese population, which has been unpublished before and suggested physicians to pay more attention to individuals with higher apoB levels as they might have higher risk of early-onset atherosclerotic stroke. In addition, since the measurements of apolipoproteins was internationally standardized and easily automated [12], the conclusions of the present study showed more promising translational potential to be tested or intervened in stroke prevention compared with those of other studies on factors influencing the onset age of stroke, such as Jerrard-Dunne’s and Kim’s studies that reported family history of early stroke [13] and the rs11833579 A/A or G/A genotype [14] as predictors of early-onset stroke, respectively. In AMORIS study focusing on cardiovascular disease, authors have presumed that the effects of certain risk factor on the disease could be determined as the function of time when the artery is exposed and its magnitude [15]. This conclusion might also fit in atherosclerotic stroke as they share similar pathophysiological mechanisms. Hence, for patients with early-onset stroke in our study, the effect of increased apoB levels outweighed the decrease effect of shorter exposure. Late-onset patients with lower apoB levels but longer exposure time reached the same outcome. Moreover, the insignificant association between the second tertile of C
2014 Informa Healthcare USA, Inc.
189
apoB levels and middling-onset group might be interpreted as the effect of increased apoB levels counteracting that of longer exposure time. Thus, the imperfection of our data in a sort of sense supported the conclusions. In the present study, the association between increased apoB/A1 ratio and early-onset stroke was weaker and less significant than that of apoB levels. These results indicated apoB was the primary drive of the results in our population, which was consistent with some previous cases [5,16,17]. However, results of other studies have suggested that apoA1 contributes more to the results than apoB [6,18,19]. This disparity might result from the difference of target populations and study designs and indicated that it might be more appropriate for subsequent studies about apolipoproteins to provide detailed data about apoB and apoA1, respectively, and then investigate which one of the apolipoprotein profiles contributes the most to the results, such as how the present study and the INTERSTROKE study mentioned above handled the issue. The latter one displays detailed data of the apolipoprotein profile, which suggests apoA1 contributes more to the results than apoB [5]. This methodology might provide more information on the pathophysiological mechanisms of the research target as apoB reflects the total number of atherogenic particles while ApoA1 has a central role in the reverse cholesterol transport and reflects antiatherogenic activities [9]. Our findings must be interpreted in the light of certain limitations. First, the study has limitations in its cross-sectional nature, only allowing to find correlations between apoB levels and the onset age of stroke, but not to assess the cause–effect relationships. However, a causal relationship between apoB and atherosclerosis/stroke [5–9] might offer some proof to the cause–effect relationship between the two and also to the conclusions of the present study. Second, there might be a selection bias since our data were collected in a single center while charactering the subjects in a detailed manner. However, we took several measures to make the results more reliable, such as to analyze apoB and apoB/A1 ratio as categorical variables and adjust them for LAA and SAA groups. Also, in order to consolidate the associations identified when comparing early-onset group and middle onset group with late-onset one, respectively, early-onset group was compared with middle- and late-onset groups combined and the results were proved to be stable.
Conclusions Our study demonstrated that the higher the apoB levels were, the earlier an atherosclerotic stroke might occur. The conclusion should be extrapolated prudently
190
F. Ye et al.
that physicians need to pay more attention to individuals with high apoB levels and take individualized strategies to prevent early-onset atherosclerotic stroke, which is in absolute need to be further confirmed in populationbased prospective studies.
8.
9.
Declaration of Interest The authors declare that they have no conflict of interest. The authors alone are responsible for the content and writing of the paper.
10.
11.
Funding 12.
Sources of support are none.
References 13. 1. Mathers CD, Boerma T, Fat DM. Global and regional causes of death. Br Med Bull 2009;92:7–32. 2. Feigin VL, Forouzanfar MH, Krishnamurthi R, et al. Global and regional burden of stroke during 1990–2010: findings from the Global Burden of Disease Study 2010. Lancet 2014;383:245–54. 3. Brown DL, Boden-Albala B, Langa KM, et al. Projected costs of ischemic stroke in the United States. Neurology 2006;67:1390–5. 4. Kang HY, Lim SJ, Suh HS, Liew D. Estimating the lifetime economic burden of stroke according to the age of onset in South Korea: a cost of illness study. BMC Public Health 2011;11:646. 5. O’Donnell MJ, Xavier D, Liu L, et al. Risk factors for ischaemic and intracerebral haemorrhagic stroke in 22 countries (the INTERSTROKE study): a case-control study. Lancet 2010;376(9735):112–23. 6. Canou¨ı-Poitrine F, Luc G, Bard JM, et al. Relative contribution of lipids and apolipoproteins to incident coronary heart disease and ischemic stroke: the PRIME Study. Cerebrovasc Dis 2010;30(3):252–9. 7. Holme I, Aastveit AH, Hammar N, et al. Relationships between lipoprotein components and risk of ischaemic and haemorrhagic
14.
15. 16.
17.
18.
19.
stroke in the Apolipoprotein MOrtality RISk study (AMORIS). J Intern Med 2009;265:275–87. Koren-Morag N, Goldbourt U, Graff E, Tanne D. Apolipoproteins B and A1 and the risk of ischemic cerebrovascular events in patients with preexisting atherothrombotic disease. J Neurol Sci 2008;270:82–7. Sniderman AD, Faraj M. Apolipoprotein B, apolipoprotein A-I, insulin resistance and the metabolic syndrome. Curr Opin Lipidol 2007;18:633–7. Walldius G, Jungner I. Apolipoprotein B and apolipoprotein A-I: risk indicators of coronary heart disease and targets for lipid-modifying therapy. J Intern Med 2004;255:188– 205. Schoenborn CA, Adams PE. Health behaviors of adults: United States, 2005–2007. Vital Health Stat 2010;10(245):80–1; Appendix II. Marcovina SM, Albers JJ, Kennedy H, et al. International Federation of Clinical Chemistry standardization project for measurements of apolipoproteins A-I and B. IV. Comparability of apolipoprotein B values by use of International Reference Material. Clin Chem 1994;40(4):586–92. Jerrard-Dunne P, Cloud G, Hassan A, Markus HS. Evaluating the genetic component of ischemic stroke subtypes: a family history study. Stroke 2003;34:1364–9. Kim DE, Noh SM, Jeong SW, Cha MH. NINJ2 SNP may affect the onset age of first-ever ischemic stroke without increasing silent cerebrovascular lesions. BMC Res Notes 2012;5: 155. Sniderman AD, Furberg CD. Age as a modifiable risk factor for cardiovascular disease. Lancet 2008;371:1547–9. Lamarche B, Moorjani S, Lupien PJ, et al. Apolipoprotein AI and B levels and the risk of ischemic heart disease during a five-year follow-up of men in the Qu´ebec cardiovascular study. Circulation 1996;94(3):273–8. Sabino AP, Sousa MDO, Lima LM, et al. ApoB/ApoA-I ratio in young patients with ischemic cerebral stroke or peripheral arterial disease. Trans Res 2008;152:113–8. Park JH, Hong KS, Lee EJ, et al. High levels of apolipoprotein B/AI ratio are associated with intracranial atherosclerotic stenosis. Stroke 2011;42:3040–6. Kostapanos MS, Christogiannis LG, Bika E, et al. Apolipoprotein B-to-A1 ratio as a predictor of acute ischemic nonembolic stroke in elderly subjects. J Stroke Cerebrovasc Dis 2010;19(6):497–502.
International Journal of Neuroscience
Copyright of International Journal of Neuroscience is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
ORIGINAL ARTICLE
Higher apolipoprotein B levels are associated with earlier onset of first-ever atherosclerotic stroke Fang Ye, Jing Liu, Shu Yang, and Fu-Qiang Guo Department of Neurology, Sichuan Provincial Academy of Medical Sciences & Sichuan Provincial People’s Hospital, Chengdu, China Background: Many studies have revealed apolipoproteins are risk factors for ischemic stroke, but the influence of apolipoproteins on onset age of first-ever atherosclerotic stroke has not been well investigated. Methods: We recruited 357 qualified participants from consecutive patients with acute ischemic stroke who came to the stroke registry center in Sichuan Provincial People’s Hospital, Chengdu, China. Patients were stratified into tertiles according to the distributions of apoB levels for large artery atherosclerosis (LAA) and small artery atherosclerosis (SAA) groups. The onset age of stroke was analyzed tripartitely in terms of early-onset group, the middlingonset group and the late-onset group. Multinomial logistical regression was used to analyze the associations between the two. Results: The risk of early-onset stroke increased monotonically with higher apoB levels (the second tertile, adjusted OR = 2.61, 95% CI 1.18–5.79 (p = 0.018); the third tertile, adjusted OR = 19.52, 95% CI 5.93–64.31 (p < 0.001)), and patients with the highest tertile of apoB levels had a 9.20 times (95% CI, 2.97–28.53, p < 0.001) increased risk of middling-onset stroke in reference to late onset of stroke. Conclusions: The present study suggests the higher the apolipoprotein B levels are, the earlier an atherosclerotic stroke might occur in a Chinese population. KEYWORDS: stroke, atherosclerotic, onset age, apolipoprotein B, apolipoprotein A1, apolipoprotein B/A1 ratio
Introduction Stroke is a leading cause of mortality and disability worldwide [1]. One third of strokes occur in those below the age of 65 and cause huge burden to the family and society [2]. Study of Brown et al. shows that loss of earnings is multiplied if the stroke occurs earlier in life [3]. Another study reveals that patients with early-onset stroke account for 75% of the total national lifetime costs of stroke but only 30% of the total incidence [4]. Therefore, to explore influencing factors for early-onset stroke is of great importance on which limited studies has focused. Apolipoproteins are consider to be promising predictors for stroke [5–8]. The INTERSTROKE study, a multicenter case–control study of 2337 patients,
Received 2 June 2014; revised 30 July 2014; accepted 30 July 2014 Correspondence: Fang Ye and Fu-Qiang Guo, Department of Neurology, Sichuan Provincial Academy of Medical Sciences & Sichuan, Provincial People’s Hospital, Chengdu 610072, China. Tel: +86 28 87394782/86 28 87394093. Fax: 86 28 87394093. E-mail: [email protected]; [email protected]
186
reveals that apoB/A1 ratio is an independent risk factor of ischemic stroke and the population- attributable risk is 35.2% [5], but few studies have investigated the impact of apolipoproteins (apo) on onset age of ischemic stroke. As apolipoproteins mainly function during the process of atherosclerosis [9], the current study targeted population with first-ever atherosclerotic stroke and was designed to investigate the association between onset age of stroke and serum apolipoproteins adjusting for potential confounders.
Methods and material Patients and relevant examinations and blood tests From 1 March 2013 to 31 January 2014, we prospectively recruited consecutive patients with acute firstever atherosclerotic stroke who came to the stroke registry center in Sichuan Provincial People’s Hospital, Chengdu, China. Informed consent was obtained from all participants, and the study protocol was approved by the Institutional Ethics Committee. The diagnosis
Apolipoproteins and onset age of stroke
was made by experienced neurologists (F. Ye, J. Liu, F.Q. Guo). The inclusion criteria of patients were acute first-ever atherosclerotic stroke with focal neurological impairments and imaging of an acute clinically relevant brain lesion excluding the following: those who had strokes caused by embolism or other undetermined etiologies; those who had suffered from ischemic cerebral stroke or its equivalents such as coronary artery disease and peripheral arterial disease (identified according to clear past history supported by symptoms, signs, treatments, reliable test reports and/or medical files) before the present event; those who had been on statin or fibrate regularly before admission since these drugs affect serum apolipoprotein levels [10]; those who underwent incomplete vascular imaging and laboratory tests and those who did not provide informed consent. Clinical data were collected by trained medical staff (S. Yang) blinded to the classification of the patients. Mean blood pressure was measured by 24 h ambulatory blood pressure monitoring (ABPM) or calculated from at least 10 measures of random BP during hospitalization. Electrocardiogram and echocardiography were performed in all patients to help exclude cardiac embolism; 24 h Holter monitoring was performed when the results of electrocardiogram were abnormal or the presence of arrhythmia was suspected in order to further exclude paroxysmal atrial fibrillation. Fasting concentrations of lipid and apolipoprotein profile were assayed within 48 h from admission by enzymatic techniques (Hitachi 7060 Automatic Biochemical Analyzer, Tokyo, Japan) and immunonephelometry (Dade Behring BN ProSpec, Deerfield, MA), respectively.
Patients’ classification and definition of risk factors All patients took the examinations of non-invasive vascular imaging for both intracranial and extracranial vessels and were classified into the large artery atherosclerosis (LAA) group or the small artery atherosclerosis (SAA) group based on whether there was at least one artery with ≤50% stenosis in the intracranial or extracranial arteries. Vascular imaging findings were reviewed by two independent reviewers (F. Ye and J. Liu) who were blinded to the clinical information. Participants who were taking antihypertensive medication or newly diagnosed hypertension were defined as currently having hypertension. Individuals were classified as having diabetes mullites (DM) if individuals were taking hypoglycemic therapies or newly diagnosed DM. Smoking was defined as currently smoking at least 1 cigarette per day for at least 3 months or at least 100 cigarettes in their lifetime [11]. C
2014 Informa Healthcare USA, Inc.
187
Statistical analysis Data are summarized as mean ± SD or number of subjects (percentage). Differences between two groups were tested using t-test. The multinomial logistical regression was used to analyze associations between serum apolipoprotein levels and onset age of stroke, adjusting for other potential confounders. Results are given by odd ratios (ORs) and its 95% confidential intervals (CIs). A p value 0.05 was considered statistically significant. All analyses were conducted using SPSS Version 19.0.
Results Three hundred fifty-seven patients [mean age at stroke onset, 65.96 ± 11.96 years; male, 229 (64.14%); patients with hypertension, 310 (86.83%); DM, 185 (51.82%); LAA, 163 (45.66%); smoking, 154 (43.14%)] were enrolled from 669 consecutive patients with acute ischemic stroke. Three hundred twelve patients were excluded, among whom, 72 had prior statin or fibrate use; 54 had ischemic stroke or its equivalents before the present event; 145 had strokes caused by cardiac embolism or other less common etiology; 31 had incomplete data and 10 had no informed consent. When analyzing the onset age, three groups were divided according to tertiles of the onset age, defined as the early-onset group with onset age 73 years, respectively. Multinomial logistic regression analyses were performed and the results were showed in Table 1. Among the covariants evaluated, apoB levels as a continuous variable were the only one with significant results when comparing either early-onset or middlingonset group to late-onset one with adjusted ORs of 7.42 (95% CIs 1.70–32.38, p = 0.008) and 7.45 (95% CIs 1.45–31.76, p = 0.007), respectively. Then, the patients were stratified into tertiles according to the distributions of apoB levels or apoB/A1 ratios, respectively, for LAA and SAA groups, because there was obvious difference between the two groups in terms of apoB levels (1.22 ± 0.49 vs. 0.97 ± 0.38, p < 0.001) and apoB/A1 ratios (0.88 ± 0.39 vs. 0.68 ± 0.28, p < 0.001). Multinomial logistic regression analyses were performed among tertiles of the apolipoproteins and onset age of stroke. Table 2 shows the results. When comparing early-onset group to late-onset one, apoB levels were significantly associated with higher risk of early-onset stroke with a dose–response relationship: the second tertile, adjusted OR = 2.61, 95% CIs 1.18–5.79 (p = 0.018); the third tertile, adjusted OR = 19.52, 95% CIs 5.93–64.31 (p < 0.001), when referenced to the first
188
F. Ye et al.
Table 1.
Multinomial logistic regression analyses of apoB, apoA1, apoB/A1 ratio and tertiles of onset age. Early-onset groupa Crude OR (95%CI)
p
Adjusted OR (95%CI)
p
2.74(1.56–4.80) 2.37(1.40–4.00) 1.07(0.51–2.24) 0.97(0.58–1.61) 1.18(0.94–1.49) 1.93(1.33–2.81) 1.31(0.96–1.79) 0.23(0.10–0.51) 2.41(1.28–4.53) 0.20(0.07–0.54) 5.20(2.22–12.18)
0.000 0.001 0.851 0.897 0.151 0.001 0.092 0.000 0.006 0.002 0.000
1.65(0.78–3.48) 1.67(0.85–3.29) N/A 0.82(0.47–1.42) N/A 1.74(1.15–2.62) 0.55(0.26–1.15) 0.45(0.14–1.48) 7.42(1.70–32.38) 0.70(0.16–2.96) 5.66(1.21–26.50)
0.189 0.137
1.05(0.52–2.10) 1.31(0.66–2.61) N/A 1.60(0.93–2.74) N/A 1.20(0.78–1.84) 0.55(0.26–1.15) 0.27(0.08–0.87) 7.45(1.75–31.76) 5.00(1.17–21.30) 2.25(0.48–10.51)
0.901 0.436
Male Smoking Hypertension DM TC, 1-mmol/L difference TG, 1-mmol/L difference LDL-C, 1-mmol/L difference HDL-C, 1-mmol/L difference ApoB, 1-g/L difference ApoA1, 1-g/L difference ApoB/A1ratio, 1-unit differenceb
0.478 0.008 0.112 0.189 0.008 0.623 0.028
Middling-onset groupa Male Smoking Hypertension DM TC, 1-mmol/L difference TG, 1-mmol/L difference LDL-C, 1-mmol/L difference HDL-C, 1-mmol/L difference ApoB, 1-g/L difference ApoA1, 1-g/L difference ApoB/A1 ratio, 1-unit differenceb
1.19(0.71–1.99) 1.24(0.73–2.10) 1.25(0.59–2.67) 1.85(1.10–3.10) 1.17(0.93–1.47) 1.51(1.03–2.21) 1.38(1.01–1.88) 0.61(0.30–1.26) 2.56(1.37–4.83) 1.37(0.54–3.49) 2.87(1.22–5.74)
0.512 0.421 0.564 0.020 0.190 0.034 0.044 0.182 0.003 0.512 0.016
0.088 0.407 0.113 0.028 0.007 0.030 0.304
Bold and italic values indicate those with a significant p value. Apo indicates apolipoprotein; LAA, large artery atherosclerosis; SAA, small artery atherosclerosis; OR, odds ratio; CI, confidence interval; DM, Diabetes mellitus; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; TG, triglyceride; TC, cholesterol. a Compared with late-onset group. b ApoB/A1 ratio and serum levels of apoB and apoA1 entered the multivariate analyses separately.
dicated 1.26 times of that (95% CIs, 0.59–2.70, p = 0.550). For either the second or third tertile of apoB levels, the risk of early-onset stroke was higher than that of middle-onset stroke. As for apoB/A1 ratio, there was significant association between the third tertile of apoB/A1
tertile. When comparing middling-onset group to lateonset one, the dose–response relationship still existed as the third tertile of apoB levels indicated 9.20 times (95% CIs, 2.97–28.53, p < 0.001) increased risk of middlingonset stroke, and the second tertile of apoB levels in-
Table 2.
Logistic regression analyses of tertiles of apoB and apoB/A1 ratio and that of onset age (total 357 patients). Middling and late-onset groups as referent group
Late-onset group as referent group Early-onset group a
Adjusted OR (95% CIs)
Middling-onset group p
1st 2nd 3rd
Referent 1.44 (0.68–3.04) 4.26 (1.55–11.71)
0.346 0.005
1st 2nd 3rd
Referent 2.61 (1.18–5.79) 19.52 (5.93–64.31)
0.018 0.000
a
Adjusted OR (95% CIs) Tertiles of apoB/A1 ratiob Referent 0.96 (0.48–1.92) 2.33 (0.90–6.00) Tertiles of apoB levelsc Referent 1.26 (0.59–2.70) 9.20 (2.97–28.53)
Early-onset group p
Adjusted ORa (95% CIs)
p
0.902 0.081
Referent 1.48 (0.75–2.90) 2.72 (1.17–6.32)
0.254 0.020
0.550 0.000
Referent 2.20 (1.09–4.44) 5.02 (1.92–13.12)
0.028 0.001
Bold and italic values indicate those with a significant p value. a Adjusted for confounders listed in Table 1. b Ranges of apoB/A1 ratio tertile for LAA and SAA group were 0.98 and 0.77, respectively. c Ranges of tertiles of apoB levels for LAA and SAA group were 1.38 g/L and 1.08 g/L, respectively.
International Journal of Neuroscience
Apolipoproteins and onset age of stroke
ratio and increased risk of early-onset stroke (adjusted OR = 4.26, 95% CIs 1.55–11.71, p = 0.005) while negative results were found in the other three comparisons. When comparing early-onset group with middle- and late-onset groups combined, dose–response relationship was revealed for apoB levels with significance (the second tertile, adjusted OR = 2.20, 95% CIs 1.09–4.44, p = 0.028; the third tertile, adjusted OR = 5.02, 95% CIs 1.92–13.12, p = 0.001, in reference to the first tertile) but not for apoB/A1 ratios.
Discussion Our study demonstrated that the risk of early-onset stroke increased monotonically with higher apoB levels (approximately 3 times for the second tertile and 20 times for the third one), and patients with the highest tertile of the apoB levels had a nine-fold increased risk of middling-onset stroke in reference to late onset of stroke. For either the second or third tertile of apoB levels, the risk of early-onset stroke were higher than that of middle-onset stroke. These results indicated that the magnitude of the associations between tertiles of apoB levels and that of onset age was not only dose-responsive, but also age-related (inversely). Thus, a conclusion might be drawn that higher apoB levels might be associated with earlier onset of first-ever acute atherosclerotic stroke in a Chinese population, which has been unpublished before and suggested physicians to pay more attention to individuals with higher apoB levels as they might have higher risk of early-onset atherosclerotic stroke. In addition, since the measurements of apolipoproteins was internationally standardized and easily automated [12], the conclusions of the present study showed more promising translational potential to be tested or intervened in stroke prevention compared with those of other studies on factors influencing the onset age of stroke, such as Jerrard-Dunne’s and Kim’s studies that reported family history of early stroke [13] and the rs11833579 A/A or G/A genotype [14] as predictors of early-onset stroke, respectively. In AMORIS study focusing on cardiovascular disease, authors have presumed that the effects of certain risk factor on the disease could be determined as the function of time when the artery is exposed and its magnitude [15]. This conclusion might also fit in atherosclerotic stroke as they share similar pathophysiological mechanisms. Hence, for patients with early-onset stroke in our study, the effect of increased apoB levels outweighed the decrease effect of shorter exposure. Late-onset patients with lower apoB levels but longer exposure time reached the same outcome. Moreover, the insignificant association between the second tertile of C
2014 Informa Healthcare USA, Inc.
189
apoB levels and middling-onset group might be interpreted as the effect of increased apoB levels counteracting that of longer exposure time. Thus, the imperfection of our data in a sort of sense supported the conclusions. In the present study, the association between increased apoB/A1 ratio and early-onset stroke was weaker and less significant than that of apoB levels. These results indicated apoB was the primary drive of the results in our population, which was consistent with some previous cases [5,16,17]. However, results of other studies have suggested that apoA1 contributes more to the results than apoB [6,18,19]. This disparity might result from the difference of target populations and study designs and indicated that it might be more appropriate for subsequent studies about apolipoproteins to provide detailed data about apoB and apoA1, respectively, and then investigate which one of the apolipoprotein profiles contributes the most to the results, such as how the present study and the INTERSTROKE study mentioned above handled the issue. The latter one displays detailed data of the apolipoprotein profile, which suggests apoA1 contributes more to the results than apoB [5]. This methodology might provide more information on the pathophysiological mechanisms of the research target as apoB reflects the total number of atherogenic particles while ApoA1 has a central role in the reverse cholesterol transport and reflects antiatherogenic activities [9]. Our findings must be interpreted in the light of certain limitations. First, the study has limitations in its cross-sectional nature, only allowing to find correlations between apoB levels and the onset age of stroke, but not to assess the cause–effect relationships. However, a causal relationship between apoB and atherosclerosis/stroke [5–9] might offer some proof to the cause–effect relationship between the two and also to the conclusions of the present study. Second, there might be a selection bias since our data were collected in a single center while charactering the subjects in a detailed manner. However, we took several measures to make the results more reliable, such as to analyze apoB and apoB/A1 ratio as categorical variables and adjust them for LAA and SAA groups. Also, in order to consolidate the associations identified when comparing early-onset group and middle onset group with late-onset one, respectively, early-onset group was compared with middle- and late-onset groups combined and the results were proved to be stable.
Conclusions Our study demonstrated that the higher the apoB levels were, the earlier an atherosclerotic stroke might occur. The conclusion should be extrapolated prudently
190
F. Ye et al.
that physicians need to pay more attention to individuals with high apoB levels and take individualized strategies to prevent early-onset atherosclerotic stroke, which is in absolute need to be further confirmed in populationbased prospective studies.
8.
9.
Declaration of Interest The authors declare that they have no conflict of interest. The authors alone are responsible for the content and writing of the paper.
10.
11.
Funding 12.
Sources of support are none.
References 13. 1. Mathers CD, Boerma T, Fat DM. Global and regional causes of death. Br Med Bull 2009;92:7–32. 2. Feigin VL, Forouzanfar MH, Krishnamurthi R, et al. Global and regional burden of stroke during 1990–2010: findings from the Global Burden of Disease Study 2010. Lancet 2014;383:245–54. 3. Brown DL, Boden-Albala B, Langa KM, et al. Projected costs of ischemic stroke in the United States. Neurology 2006;67:1390–5. 4. Kang HY, Lim SJ, Suh HS, Liew D. Estimating the lifetime economic burden of stroke according to the age of onset in South Korea: a cost of illness study. BMC Public Health 2011;11:646. 5. O’Donnell MJ, Xavier D, Liu L, et al. Risk factors for ischaemic and intracerebral haemorrhagic stroke in 22 countries (the INTERSTROKE study): a case-control study. Lancet 2010;376(9735):112–23. 6. Canou¨ı-Poitrine F, Luc G, Bard JM, et al. Relative contribution of lipids and apolipoproteins to incident coronary heart disease and ischemic stroke: the PRIME Study. Cerebrovasc Dis 2010;30(3):252–9. 7. Holme I, Aastveit AH, Hammar N, et al. Relationships between lipoprotein components and risk of ischaemic and haemorrhagic
14.
15. 16.
17.
18.
19.
stroke in the Apolipoprotein MOrtality RISk study (AMORIS). J Intern Med 2009;265:275–87. Koren-Morag N, Goldbourt U, Graff E, Tanne D. Apolipoproteins B and A1 and the risk of ischemic cerebrovascular events in patients with preexisting atherothrombotic disease. J Neurol Sci 2008;270:82–7. Sniderman AD, Faraj M. Apolipoprotein B, apolipoprotein A-I, insulin resistance and the metabolic syndrome. Curr Opin Lipidol 2007;18:633–7. Walldius G, Jungner I. Apolipoprotein B and apolipoprotein A-I: risk indicators of coronary heart disease and targets for lipid-modifying therapy. J Intern Med 2004;255:188– 205. Schoenborn CA, Adams PE. Health behaviors of adults: United States, 2005–2007. Vital Health Stat 2010;10(245):80–1; Appendix II. Marcovina SM, Albers JJ, Kennedy H, et al. International Federation of Clinical Chemistry standardization project for measurements of apolipoproteins A-I and B. IV. Comparability of apolipoprotein B values by use of International Reference Material. Clin Chem 1994;40(4):586–92. Jerrard-Dunne P, Cloud G, Hassan A, Markus HS. Evaluating the genetic component of ischemic stroke subtypes: a family history study. Stroke 2003;34:1364–9. Kim DE, Noh SM, Jeong SW, Cha MH. NINJ2 SNP may affect the onset age of first-ever ischemic stroke without increasing silent cerebrovascular lesions. BMC Res Notes 2012;5: 155. Sniderman AD, Furberg CD. Age as a modifiable risk factor for cardiovascular disease. Lancet 2008;371:1547–9. Lamarche B, Moorjani S, Lupien PJ, et al. Apolipoprotein AI and B levels and the risk of ischemic heart disease during a five-year follow-up of men in the Qu´ebec cardiovascular study. Circulation 1996;94(3):273–8. Sabino AP, Sousa MDO, Lima LM, et al. ApoB/ApoA-I ratio in young patients with ischemic cerebral stroke or peripheral arterial disease. Trans Res 2008;152:113–8. Park JH, Hong KS, Lee EJ, et al. High levels of apolipoprotein B/AI ratio are associated with intracranial atherosclerotic stenosis. Stroke 2011;42:3040–6. Kostapanos MS, Christogiannis LG, Bika E, et al. Apolipoprotein B-to-A1 ratio as a predictor of acute ischemic nonembolic stroke in elderly subjects. J Stroke Cerebrovasc Dis 2010;19(6):497–502.
International Journal of Neuroscience
Copyright of International Journal of Neuroscience is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
