Inter-individual differences in biological aging could affect susceptibility to stroke. To date, the relationship between stroke and telomere shortening remain inconclusive; and sparse data are available for haemorrhagic stroke. A Chinese case-control study was conducted, comprising 1756 cases (767 atherothrombosis, 503 lacunar infarction and 486 haemorrhagic strokes) and 1801 controls. Stroke patients were prospectively followed up for a median of 4.5 (range, 0.1–6.0) years. Individuals with shorter telomere length had a higher presence of atherothrombotic stroke {multivariate OR (odds ratio) 1.37 [95% CI (confidence interval), 1.06–1.77]; P=0.015} or haemorrhagic stroke [multivariate OR 1.48 (95% CI, 1.08–2.02); P=0.016] in comparison of the lowest to highest tertile of telomere length. Particularly, in subjects with a family history of stroke, there was a significant 2.55-fold increased presence of atherothrombotic stroke (95% CI, 1.87–3.48; Ptrend<0.0001) and a 2.33-fold increased presence of haemorrhagic stroke (95% CI, 1.62–3.36; Ptrend<0.0001). During the follow-up, 338 recurrent strokes and 312 deaths (181 from stroke or coronary heart disease and 131 from other causes) were documented. Associations with stroke recurrence were not observed in the follow-up patients, whereas atherothrombotic stroke cases with shorter telomeres had 69% increased risk of post-stroke death [relative risk, 1.69 (95% CI, 1.07–2.67); P=0.02]. Finally, we compared telomere lengths in 12 paired samples of circulating leucocytes and carotid atherosclerotic plaques from patients undergoing carotid endarterectomy; there was a positive correlation between vessel wall tissue and leucocyte telomere length. In conclusion, shorter telomere length may serve as a potential marker for the presence of atherothrombotic and haemorrhagic stroke and for the risk of post-stroke death.

CLINICAL PERSPECTIVES

  • Inter-individual differences in biological aging could affect susceptibility to stroke, and telomere attrition represents one molecular mechanism that contributes to cellular aging. However, the association of leucocyte telomere length with presence of ischaemic stroke and haemorrhagic stroke has not been investigated.

  • Our data from this large Chinese stroke population provided evidence that shorter leucocyte telomeres are not only associated with the presence of atherothrombotic and haemorrhagic stroke, particularly in those individuals with a family history of stroke, but also independently predict worse prognosis of death after the occurrence of atherothrombotic stroke during the long-term follow-up.

  • Additional analyses, by using paired samples of blood leucocyte and human carotid atherosclerotic plaques, justify the use of blood leucocyte telomeres as an effective surrogate for vascular aging and related diseases.

INTRODUCTION

Stroke is a major cause of death and disability worldwide. Survivors are often disabled, require long-term care, and are at increased risk of recurrence. Although traditional vascular risk factors potentially explain most of the risk of stroke, at an individual level there is a wide variation in the development of stroke and clinical manifestation, even in individuals with similar risk factor profiles. The reasons why there are inter-individual heterogeneities in stroke clinical phenotypes are still incompletely understood, but biological aging is one of major contributors to stroke [1].

Telomere attrition represents one molecular mechanism that contributes to cellular aging. Telomeres are specialized DNA–protein structures at the end of all chromosomes, which preserve chromosome stability and integrity [2]. When telomere lengths are shortened to a critical value, they lose capping function at the chromosomal ends, resulting in activation of DNA damage checkpoints. Emerging evidence shows that telomere attrition is linked to vascular endothelial cell senescence and apoptosis and may contribute to atherogenesis [3].

The association between stroke and telomere shortening has been addressed; however, the results remain inconsistent. Prospective studies from the U.S.A. have found no association between leucocyte telomere length and stroke [4] or death from cardiovascular events including stroke [5]. Another study shows that leucocyte telomere length is associated with cardiovascular outcomes but not stroke in elderly high-risk hypertensive patients from the European population [6]. A positive association between shorter telomere length and ischaemic stroke and mortality from all causes has recently been found in the Chinese population by Ding et al. [7]. However, to date, most studies are limited to ischaemic stroke, with less information available on haemorrhagic stroke.

Shorter telomeres reflect increased leucocyte turnover rate as the burden of chronic inflammation and oxidative stress accumulate during the individual's life course [8]. Blood leucocytes are usually used to study cellular aging as measured by telomere length because of their accessibility in routine clinical studies [9], but it remains to be established whether leucocyte telomere length in blood is an appropriate index of vascular telomere length in diseased vessel wall such as human atherosclerotic plaque lesions.

In the present study, we investigated the associations of leucocyte telomere length with the presence of ischaemic stroke and haemorrhagic stroke in a large case-control study in the Chinese Han population. We further assessed the relationship between telomere length and subsequent prognosis of stroke by prospectively following up the patients with stroke for a median 4.5 years. The correlation between telomere length of blood leucocytes and that of the vascular wall in human carotid atherosclerotic plaques was also evaluated.

MATERIALS AND METHODS

Stroke sample population

The Multicenter Chinese Stroke Study comprised 2000 consecutive stroke patients (age, 35–74 years) and 2000 age- and gender-matched control subjects, which has been described previously [10]. The criteria of diagnosis and recruitment, measurement of plasma biochemical variables and collection of clinical characteristics are given in detail in the Supplementary Materials and methods section (available at http://www.clinsci.org/cs/125/cs1250027add.htm). In brief, stroke was defined as the sudden onset of a non-convulsive and focal neurological deficit persisting for over 24 h, and confirmed by strict neurological examination, CT (computed tomography) or MRI (magnetic resonance imaging) according to the International Classification of Diseases (9th Revision). We recruited three subtypes of stroke: cerebral thrombosis (atherothrombosis), lacunar infarction (lacunar) and ICH (intracerebral haemorrhage). The enrolled patients were survivors of acute stroke events. Other types of stroke (embolic stroke and subarachnoid haemorrhage) and systemic diseases (collagenosis, inflammation, liver or renal diseases) were excluded.

Before data assessment, we excluded 443 subjects because of lack of complete clinical records (95 cases and 72 control subjects) and insufficient DNA (149 cases and 127 control subjects). No significant differences were found in clinical characteristics between the included and the excluded subjects. Finally, 1756 cases (767 atherothrombotic strokes, 503 lacunar infarction and 486 haemorrhagic strokes) and 1801 control subjects were eligible for the analysis using a case-control approach. The study was approved by the ethics committees of collaborating hospitals. All participants reported themselves as Han nationality, and provided written informed consent.

Follow-up and outcome assessments

Stroke patients in this cohort were prospectively followed up over a 2-year period until 31 May 2006, with a standard questionnaire by using telephone contact by physician investigators. The endpoints were stroke recurrence and post-stroke deaths. Recurrent stroke was defined using the following criteria: clinical evidence of the sudden onset of a new focal neurological deficit with no apparent cause other than that of vascular origin occurring at any time after the index stroke; or, there was clinical evidence of the sudden onset of an exacerbation of a previous focal neurological deficit with no apparent cause other than that of vascular origin occurring >21 days after the index stroke [11]. All hospitalized patients were confirmed by a local neurologist, based on the direct review of patient medical records and brain imaging (CT or MRI). The non-hospitalized patients were evaluated by primary care physicians in the community. Deaths were reported by family members, work associates and/or obtained from death certificates and medical records.

Telomere length measurement

Genomic DNA was isolated from peripheral blood leucocytes according to standard procedures. Relative mean leucocyte telomere length was determined with a quantitative real-time PCR-based technique that compares telomere repeat copy number (T) to single-copy gene copy number (S) (T/S ratio) in a given sample [12]. All PCRs were performed in triplicate on an ABI PRISM 7900HT Fast real-time PCR system (Applied Biosystems). In brief, 10 ng of genomic DNA was dried in a 384-well plate and then resuspended in 10 μl of either the telomere or single-copy gene (the human β-globin gene on chromosome 11p15.5) PCR mixture. The primer sequences and thermal cycling profiles are given in detail in the Supplementary Materials and methods section. A reference calibrator sample was included with each measurement to control inter-assay variability. The average inter-plate coefficient of variability for the telomere and β-globin assays was 6.4 and 4.6%, respectively. A standard curve derived from serially diluted reference DNA (1.56–100 ng; 2-fold dilution; seven points) was run for both the telomere and the β-globin PCRs, with good linearity (R2>0.98). As part of routine quality control, 10% of the samples were randomly chosen to test the reproducibility of the assay. All measurements were performed by laboratory personnel blinded to the case-control status and the outcome assessment.

Statistical analysis

Normal distribution of data was examined by the Kolmogorov–Smirnov normality test. The T/S ratios of mean leucocyte telomere length were natural logarithm transformed because of a skewed distribution. Characteristics of cases and controls were compared by ϕ2 tests for categorical variables and ANOVA for quantitative variables. The distributions of the T/S ratio of telomere length were divided into tertiles among the control subjects, and the cut-off values were <0.340 for the lowest tertile, 0.340–0.826 for the middle tertile and >0.826 for the highest tertile. The ORs (odds ratio) and 95% CIs (confidence intervals) were estimated for the association between telomeres and the presence of stroke by using logistic regression models. We performed separate data analysis, interpretation and presentation on the three subtypes of atherothrombotic stroke, lacunar infarction and hemorrhagic stroke, because they have different pathophysiology. Multivariate analyses were first adjusted for age and gender, and further for BMI (body mass index), fasting triacylglycerols, total cholesterol, HDL (high-density lipoprotein)-cholesterol), fasting blood glucose, BP (blood pressure), smoking status (never, past, current), alcohol intake (current drinker, yes/no), history of hypertension (yes/no), diabetes (yes/no), previous cardiovascular disease (yes/no), family history of stroke (yes/no), and medication treatment (yes/no). The significance of multiplicative interactions between the telomeres and family history of stroke was determined by the likelihood-ratio test.

Person-years of follow-up started from the date of diagnosis of the first-ever stroke until the date of a first recurrence of stroke, death or the end of the follow-up (May 31, 2006). For those who were lost to follow-up [94 out of 1756 (5.3%)], it ended with the date last known to be alive. Cox proportional-hazards models were used to examine the association between telomeres and the risk of stroke recurrence or mortality after adjustment for age, gender, conventional risk factors, stroke subtypes, neurological deficit and medication treatment.

To examine whether leucocyte telomere length is an appropriate index of vascular telomere length in diseased vessel wall, the correlation between telomere lengths in 12 paired samples of blood leucocyte and human carotid atherosclerotic plaques was assessed by using a general linear regression model adjustment for age and gender. All probability values are 2-sided, and P<0.05 is considered significant. Analyses were performed with SPSS software, version 13.0 (SPSS Inc.).

RESULTS

Clinical characteristics of study participants

In the present study, age, presented as mean±S.D., was 60.3±9.3 years in cases and 59.7±8.2 years in controls, and men accounted for 63.0% of cases and 57.3% of controls. As expected, stroke patients had a higher prevalence of conventional vascular risk factors, including smoking, alcohol intake, family history of stroke, hypertension and diabetes; they also had higher levels of blood glucose and triacylglycerols, and the lower level of HDL-cholesterol (Table 1).

Table 1
Clinical characteristics of the case/control study

Values are given as means±S.D., numbers (%) or medians (interquartile range). Telomere length is expressed as a relative telomere/single-copy gene (T/S) ratio. *P<0.05 and †P<0.01 compared with control subjects. P values were were obtained using the two-sample Student's t test for comparison of continuous variables, the χ2 test for categorical variables, and the Mann–Whitney U test for triacylglycerols and relative T/S ratio. LDL, low-density lipoprotein; SBP, systolic BP; DBP, diastolic BP.

  Stroke patients 
Characteristics Controls (n=1801) Atherothrombotic (n=767) Lacunar infarction (n=503) Haemorrhagic stroke (n=486) Total cases (n=1756) 
Age (years) 59.7±8.2 61.5±9.2† 60.8±8.5* 58.0±9.8* 60.3±9.3* 
Male (n1032 (57.3%) 485 (63.2%)* 318 (63.2%)* 303 (62.3%)* 1106 (63.0%)† 
BMI (kg/m224.2±3.3 24.4±3.6 24.5±3.2 24.0±3.5 24.3±3.5 
SBP (mmHg) 129±17 147±23† 142±20† 151±23† 147±23† 
DBP (mmHg) 80±10 87±13† 86±11† 92±13† 88±13† 
Serum glucose (mmol/l) 5.85±1.65 6.70±2.73† 6.38±2.53† 6.53±2.22† 6.56±2.54† 
Lipids (mmol/l)      
 Total cholesterol 4.99±1.00 4.86±1.03† 4.80±0.98† 4.54±0.99† 4.75±1.02† 
 Triacylglycerols 1.47 (1.05–2.11) 1.71 (1.20–2.51)† 1.72 (1.22–2.49)† 1.45 (1.09–1.98) 1.65 (1.17–2.37)† 
 HDL-cholesterol 1.06±0.30 0.89±0.26† 0.92±0.26† 0.89±0.32† 0.90±0.28† 
 LDL-cholesterol 3.12±0.95 3.06±0.96 2.93±0.95† 2.87±0.96† 2.97±0.96† 
Serum creatinine (μmol/l) 95.6±22.9 98.2±21.2† 99.1±28.0† 103.5±51.5† 99.8±33.8† 
Cigarette smoking (n) 665 (36.9%) 391 (51.0%)† 225 (45.7%)* 236 (48.6%)† 852 (48.5%)† 
Alcohol intake (n561 (31.1%) 293 (38.2%)* 168 (33.4%) 202 (41.6%)† 663 (37.8%)† 
Medical history (n     
 Hypertension 477 (26.5%) 488 (63.6%)† 299 (59.4%)† 314 (64.6%)† 1101 (62.7%)† 
 Diabetes mellitus 92 (5.1%) 125 (16.3%)† 65 (12.9%)† 25 (5.1%) 215 (12.2%)† 
 CHD 219 (12.2%) 157 (20.5%)† 58 (11.5%) 44 (9.1%) 259 (14.7%)* 
Medication (n397 (22.0%) 158 (20.6%) 122 (24.3%) 112 (23.1%) 392 (22.3%) 
Family history of stroke (n432 (24.0%) 247 (32.2%)† 147 (29.2%)* 156 (32.1%)† 550 (31.3%)† 
Leucocyte telomere length 0.53 (0.28–1.04) 0.41 (0.22–0.83)† 0.46 (0.26–0.92) 0.45 (0.22–0.83)† 0.44 (0.23–0.86)† 
  Stroke patients 
Characteristics Controls (n=1801) Atherothrombotic (n=767) Lacunar infarction (n=503) Haemorrhagic stroke (n=486) Total cases (n=1756) 
Age (years) 59.7±8.2 61.5±9.2† 60.8±8.5* 58.0±9.8* 60.3±9.3* 
Male (n1032 (57.3%) 485 (63.2%)* 318 (63.2%)* 303 (62.3%)* 1106 (63.0%)† 
BMI (kg/m224.2±3.3 24.4±3.6 24.5±3.2 24.0±3.5 24.3±3.5 
SBP (mmHg) 129±17 147±23† 142±20† 151±23† 147±23† 
DBP (mmHg) 80±10 87±13† 86±11† 92±13† 88±13† 
Serum glucose (mmol/l) 5.85±1.65 6.70±2.73† 6.38±2.53† 6.53±2.22† 6.56±2.54† 
Lipids (mmol/l)      
 Total cholesterol 4.99±1.00 4.86±1.03† 4.80±0.98† 4.54±0.99† 4.75±1.02† 
 Triacylglycerols 1.47 (1.05–2.11) 1.71 (1.20–2.51)† 1.72 (1.22–2.49)† 1.45 (1.09–1.98) 1.65 (1.17–2.37)† 
 HDL-cholesterol 1.06±0.30 0.89±0.26† 0.92±0.26† 0.89±0.32† 0.90±0.28† 
 LDL-cholesterol 3.12±0.95 3.06±0.96 2.93±0.95† 2.87±0.96† 2.97±0.96† 
Serum creatinine (μmol/l) 95.6±22.9 98.2±21.2† 99.1±28.0† 103.5±51.5† 99.8±33.8† 
Cigarette smoking (n) 665 (36.9%) 391 (51.0%)† 225 (45.7%)* 236 (48.6%)† 852 (48.5%)† 
Alcohol intake (n561 (31.1%) 293 (38.2%)* 168 (33.4%) 202 (41.6%)† 663 (37.8%)† 
Medical history (n     
 Hypertension 477 (26.5%) 488 (63.6%)† 299 (59.4%)† 314 (64.6%)† 1101 (62.7%)† 
 Diabetes mellitus 92 (5.1%) 125 (16.3%)† 65 (12.9%)† 25 (5.1%) 215 (12.2%)† 
 CHD 219 (12.2%) 157 (20.5%)† 58 (11.5%) 44 (9.1%) 259 (14.7%)* 
Medication (n397 (22.0%) 158 (20.6%) 122 (24.3%) 112 (23.1%) 392 (22.3%) 
Family history of stroke (n432 (24.0%) 247 (32.2%)† 147 (29.2%)* 156 (32.1%)† 550 (31.3%)† 
Leucocyte telomere length 0.53 (0.28–1.04) 0.41 (0.22–0.83)† 0.46 (0.26–0.92) 0.45 (0.22–0.83)† 0.44 (0.23–0.86)† 

The recruited stroke patients were prospectively followed up for a median of 4.5 (range, 0.1–6.0) years; 94 (5.3%) of 1756 cases were lost because of emigration. No substantial differences were found in baseline characteristics and T/S ratio of telomere length between the follow-up and lost-to-follow-up subjects (Supplementary Table S1 at http://www.clinsci.org/cs/125/cs1250027add.htm)]. Table 2 shows baseline characteristics of follow-up cases with ischaemic or hemorrhagic stroke by tertiles of telomere length. During the follow-up, a total of 338 recurrent strokes and 312 deaths from all causes [181 from stroke or CHD (coronary heart disease) and 131 from other causes] were documented. In the present study, 85.5% of 338 recurrent strokes were admitted to a hospital for evaluation and had a CT scan or MR imaging within 10 days of symptom onset, and 14.5% of patients were not hospitalized and were evaluated by primary care physicians in the community.

Table 2
Baseline characteristics of follow-up cases by tertiles of leucocyte telomere length in the prospective stroke cohort (n=1662)

Results are given as means±S.D., numbers (percentage) or medians (interquartile range). Telomere length is expressed as a relative telomere/single-copy gene (T/S) ratio. P values were obtained by ANOVA for comparison of continuous variables, the χ2 test for categorical variables, and the Kruskal–Wallis H test for triacylglycerols. LDL, low-density lipoprotein; SBP, systolic BP; DBP, diastolic BP.

 Atherothrombotic stroke (n=725) Lacunar infarction (n=481) Haemorrhagic stroke (n=456) 
Characteristics Highest tertile (>0.826) Middle tertile (0.340–0.826) Lowest tertile (<0.340) P value Highest tertile (>0.826) Middle tertile (0.340–0.826) Lowest tertile (<0.340) P value Highest tertile (>0.826) Middle tertile (0.340–0.826) Lowest tertile (<0.340) P value 
Participants (n181 239 305  140 157 184  110 167 179  
Age (years) 60.9±10.0 61.3±8.4 62.1±9.0 0.33 59.7±9.0 61.8±8.1 60.6±8.2 0.10 58.4±10.0 57.4±10.1 57.9±9.4 0.71 
Male (n123 (68.0%) 157 (65.7) 177 (58.0%) 0.06 86 (61.4%) 105 (66.9%) 114 (62.0%) 0.54 68 (61.8%) 113 (67.7%) 110 (61.5%) 0.43 
BMI (kg/m224.0±3.5 24.7±3.6 24.3±3.7 0.16 24.5±3.6 24.5±3.0 24.6±3.1 0.93 23.7±3.5 24.2±3.5 24.2±3.7 0.41 
SBP (mmHg) 145±22 146±22 148±24 0.25 140±21 143±18 144±19 0.14 148±22 149±23 153±23 0.19 
DBP (mmHg) 86±13 87±12 87±13 0.43 85±12 85±11 87±11 0.06 90±11 90±14 91±13 0.06 
Glucose (mmol/l) 6.64±2.84 7.74±2.57 6.61±2.7 0.86 6.35±2.40 6.43±2.88 6.34±2.23 0.93 6.62±2.41 6.22±1.71 6.79±2.54 0.06 
Lipids (mmol/l)             
 Total cholesterol 4.77±1.00 4.86±1.00 4.92±1.09 0.32 4.82±1.02 4.77±0.93 4.85±1.00 0.75 4.50±1.00 4.55±0.97 4.57±1.02 0.84 
 Triacylglycerols 1.72 (1.12–2.51) 1.71 (1.27–2.54) 1.68 (1.22–2.52) 0.75 1.70 (1.15–2.33) 1.77 (0.92–2.54) 1.71 (1.24–2.68) 0.36 1.44 (1.09–2.00) 1.56 (1.15–2.08) 1.41 (1.09–1.95) 0.52 
 HDL-cholesterol 0.92±0.27 0.87±0.25 0.88±0.26 0.20 0.93±0.24 0.96±0.30 0.88±0.22 0.007 0.88±0.28 0.93±0.40 0.85±0.27 0.13 
 LDL-cholesterol 2.95±0.07 3.06±0.94 3.13±1.03 0.14 2.99±0.93 2.84±0.92 3.00±1.01 0.29 2.81±1.07 2.84±0.80 2.93±1.04 0.55 
Creatinine (μmol/l) 97.7±20.3 98.3±20.1 97.5±22.3 0.91 95.6±26.3 100.4±27.3 100.7±30.2 0.28 103.8±22.8 102.5±30.5 105.6±30.5 0.87 
Smoking (n97 (53.6%) 120 (50.2%) 153 (50.2%) 0.73 65 (46.4%) 71 (45.2%) 78 (42.4%) 0.75 56 (50.9%) 89 (53.3%) 78 (43.6%) 0.17 
Alcohol intake (n73 (40.3%) 99 (41.4%) 105 (34.4%) 0.20 49 (35.0%) 51 (32.5%) 59 (32.1%) 0.84 44 (40.0%) 76 (45.5%) 76 (42.5%) 0.65 
Medical history (n            
 Hypertension 110 (60.8%) 156 (65.3%) 195 (63.9%) 0.63 77 (55.0%) 98 (62.4%) 114 (62.0%) 0.34 63 (57.3%) 108 (64.7%) 118 (65.9%) 0.30 
 Diabetes mellitus 24 (13.3%) 41 (17.2%) 56 (18.4%) 0.34 22 (15.7%) 13 (8.3%) 28 (15.2%) 0.09 4 (3.6%) 6 (3.6%) 14 (7.8%) 0.15 
 CHD 34 (18.8%) 51 (21.3%) 69 (22.6%) 0.61 18 (12.9%) 16 (10.2%) 23 (12.5%) 0.73 12 (10.9%) 17 (10.2%) 12 (6.7%) 0.38 
Family history of stroke (n55 (30.4%) 78 (32.6%) 104 (34.1%) 0.70 45 (32.1%) 46 (29.3%) 53 (28.8%) 0.79 31 (28.2%) 54 (32.3%) 61 (34.1%) 0.58 
 Atherothrombotic stroke (n=725) Lacunar infarction (n=481) Haemorrhagic stroke (n=456) 
Characteristics Highest tertile (>0.826) Middle tertile (0.340–0.826) Lowest tertile (<0.340) P value Highest tertile (>0.826) Middle tertile (0.340–0.826) Lowest tertile (<0.340) P value Highest tertile (>0.826) Middle tertile (0.340–0.826) Lowest tertile (<0.340) P value 
Participants (n181 239 305  140 157 184  110 167 179  
Age (years) 60.9±10.0 61.3±8.4 62.1±9.0 0.33 59.7±9.0 61.8±8.1 60.6±8.2 0.10 58.4±10.0 57.4±10.1 57.9±9.4 0.71 
Male (n123 (68.0%) 157 (65.7) 177 (58.0%) 0.06 86 (61.4%) 105 (66.9%) 114 (62.0%) 0.54 68 (61.8%) 113 (67.7%) 110 (61.5%) 0.43 
BMI (kg/m224.0±3.5 24.7±3.6 24.3±3.7 0.16 24.5±3.6 24.5±3.0 24.6±3.1 0.93 23.7±3.5 24.2±3.5 24.2±3.7 0.41 
SBP (mmHg) 145±22 146±22 148±24 0.25 140±21 143±18 144±19 0.14 148±22 149±23 153±23 0.19 
DBP (mmHg) 86±13 87±12 87±13 0.43 85±12 85±11 87±11 0.06 90±11 90±14 91±13 0.06 
Glucose (mmol/l) 6.64±2.84 7.74±2.57 6.61±2.7 0.86 6.35±2.40 6.43±2.88 6.34±2.23 0.93 6.62±2.41 6.22±1.71 6.79±2.54 0.06 
Lipids (mmol/l)             
 Total cholesterol 4.77±1.00 4.86±1.00 4.92±1.09 0.32 4.82±1.02 4.77±0.93 4.85±1.00 0.75 4.50±1.00 4.55±0.97 4.57±1.02 0.84 
 Triacylglycerols 1.72 (1.12–2.51) 1.71 (1.27–2.54) 1.68 (1.22–2.52) 0.75 1.70 (1.15–2.33) 1.77 (0.92–2.54) 1.71 (1.24–2.68) 0.36 1.44 (1.09–2.00) 1.56 (1.15–2.08) 1.41 (1.09–1.95) 0.52 
 HDL-cholesterol 0.92±0.27 0.87±0.25 0.88±0.26 0.20 0.93±0.24 0.96±0.30 0.88±0.22 0.007 0.88±0.28 0.93±0.40 0.85±0.27 0.13 
 LDL-cholesterol 2.95±0.07 3.06±0.94 3.13±1.03 0.14 2.99±0.93 2.84±0.92 3.00±1.01 0.29 2.81±1.07 2.84±0.80 2.93±1.04 0.55 
Creatinine (μmol/l) 97.7±20.3 98.3±20.1 97.5±22.3 0.91 95.6±26.3 100.4±27.3 100.7±30.2 0.28 103.8±22.8 102.5±30.5 105.6±30.5 0.87 
Smoking (n97 (53.6%) 120 (50.2%) 153 (50.2%) 0.73 65 (46.4%) 71 (45.2%) 78 (42.4%) 0.75 56 (50.9%) 89 (53.3%) 78 (43.6%) 0.17 
Alcohol intake (n73 (40.3%) 99 (41.4%) 105 (34.4%) 0.20 49 (35.0%) 51 (32.5%) 59 (32.1%) 0.84 44 (40.0%) 76 (45.5%) 76 (42.5%) 0.65 
Medical history (n            
 Hypertension 110 (60.8%) 156 (65.3%) 195 (63.9%) 0.63 77 (55.0%) 98 (62.4%) 114 (62.0%) 0.34 63 (57.3%) 108 (64.7%) 118 (65.9%) 0.30 
 Diabetes mellitus 24 (13.3%) 41 (17.2%) 56 (18.4%) 0.34 22 (15.7%) 13 (8.3%) 28 (15.2%) 0.09 4 (3.6%) 6 (3.6%) 14 (7.8%) 0.15 
 CHD 34 (18.8%) 51 (21.3%) 69 (22.6%) 0.61 18 (12.9%) 16 (10.2%) 23 (12.5%) 0.73 12 (10.9%) 17 (10.2%) 12 (6.7%) 0.38 
Family history of stroke (n55 (30.4%) 78 (32.6%) 104 (34.1%) 0.70 45 (32.1%) 46 (29.3%) 53 (28.8%) 0.79 31 (28.2%) 54 (32.3%) 61 (34.1%) 0.58 

Association between leucocyte telomere length and presence of stroke

We observed that mean leucocyte telomere length was significantly shorter in patients with stroke than in control subjects [median, 0.44 (interquartile range, 0.23–0.86) against 0.53 (interquartile range, 0.28 to 1.04) respectively; P<0.001]. Overall, the telomere length ratio was significantly inversely correlated with chronological age (see Supplementary Figure S1 at http://www.clinsci.org/cs/125/cs1250027add.htm), and decreased 9% per decade (95% CI, −3.5 to −13.8; correlation coefficient γ=−0.13; P<0.001) in control subjects and 6% per decade (95% CI, −3.2 to −14.0; correlation coefficient γ=−0.11; P<0.001) in stroke patients, respectively. However, there was no difference in the regression line slopes between controls and cases (P=0.46).

In both crude analysis and multivariable analysis adjustment for age, gender and other vascular risk factors, individuals in the lowest tertile of telomere length had a significantly higher presence of atherothrombotic stroke [multivariate OR, 1.37 (95% CI, 1.06–1.77; P=0.015] or haemorrhagic stroke [multivariate OR, 1.48 (95% CI, 1.08–2.02); P=0.016] than did those in the highest tertile (Table 3). No associations between telomere length and presence of lacunar infarction were observed. In subsidiary analyses, telomere length was analysed as a continuous variable and was found to be associated with approx. 15% increased susceptibility to atherothrombotic stroke or haemorrhagic stroke per 1 S.D. decrease in telomere length (P<0.05; Table 3). In addition, because the observed association between shorter telomeres and stroke could be biased due to history of coronary heart disease, we also performed sensitivity analyses by excluding those subjects with history of heart diseases, and the results were not substantially changed (see Supplementary Table S2 at http://www.clinsci.org/cs/125/cs1250027add.htm).

Table 3
Association between shorter leucocyte telomere length and presence of ischaemic or haemorrhagic stroke

The cut-off values of tertile of leucocyte telomere length were derived from the control group, and relative T/S ratios were <0.34 for lowest tertile, 0.34–0.826 for the middle tertile and >0.826 for the highest tertile (as the reference). OR (95% CI) was obtained with multivariate logistic regression analysis. Model I: adjustment for age, gender, BMI, systolic and diastolic BP, fasting glucose, triacylglycerols, total cholesterol, HDL-cholesterol and LDL-cholesterol. Model II: adjustment for the covariates mentioned above plus smoking status, alcohol intake, diabetes, history of hypertension, previous CHD, family history of stroke and medication.

 Leucocyte telomere length    
Group Highest tertile (>0.826) Middle tertile (0.340–0.826) Lowest tertile (<0.340) P trend Per 1−S.D. decrease in ln-transformed telomere length P value 
Controls (n=1801) 601 (33.4%) 600 (33.3%) 600 (33.3%)    
Atherothrombotic stroke (n=767) 192 (25.0%) 251 (32.7%) 324 (42.2%)    
 OR (95% CI)       
  No adjustment 1.0 1.32 (1.06–1.65) 1.69 (1.36–2.10) <0.0001 1.22 (1.12–1.33) <0.0001 
  Multivariable model I 1.0 1.20 (0.94–1.55) 1.50 (1.18–1.92) 0.003 1.19 (1.08–1.31) <0.0001 
  Multivariable model II 1.0 1.15 (0.89–1.50) 1.37 (1.06–1.77) 0.03 1.15 (1.04–1.27) 0.007 
Lacunar infarction (n=503) 148 (29.4%) 166 (33.0%) 189 (37.6%)    
 OR (95% CI)       
  No adjustment 1.0 1.12 (0.87–1.45) 1.32 (1.03–1.69) 0.09 1.12 (1.01–1.24) 0.03 
  Multivariable model I 1.0 1.04 (0.79–1.36) 1.16 (0.89–1.52) 0.52 1.07 (0.97–1.20) 0.19 
  Multivariable model II 1.0 1.04 (0.78–1.37) 1.12 (0.85–1.47) 0.72 1.06 (0.95–1.18) 0.31 
Haemorrhagic stroke (n=486) 122 (25.1%) 172 (35.4%) 192 (39.5%)    
 OR (95% CI)       
  No adjustment 1.0 1.52 (1.17–1.98) 1.63 (1.25–2.12) 0.001 1.21 (1.09–1.34) 0.001 
  Multivariable model I 1.0 1.38 (1.01–1.89) 1.49 (1.09–2.04) 0.03 1.14 (1.01–1.29) 0.03 
  Multivariable model II 1.0 1.32 (0.96–1.82) 1.48 (1.08–2.02) 0.05 1.14 (1.01–1.28) 0.04 
 Leucocyte telomere length    
Group Highest tertile (>0.826) Middle tertile (0.340–0.826) Lowest tertile (<0.340) P trend Per 1−S.D. decrease in ln-transformed telomere length P value 
Controls (n=1801) 601 (33.4%) 600 (33.3%) 600 (33.3%)    
Atherothrombotic stroke (n=767) 192 (25.0%) 251 (32.7%) 324 (42.2%)    
 OR (95% CI)       
  No adjustment 1.0 1.32 (1.06–1.65) 1.69 (1.36–2.10) <0.0001 1.22 (1.12–1.33) <0.0001 
  Multivariable model I 1.0 1.20 (0.94–1.55) 1.50 (1.18–1.92) 0.003 1.19 (1.08–1.31) <0.0001 
  Multivariable model II 1.0 1.15 (0.89–1.50) 1.37 (1.06–1.77) 0.03 1.15 (1.04–1.27) 0.007 
Lacunar infarction (n=503) 148 (29.4%) 166 (33.0%) 189 (37.6%)    
 OR (95% CI)       
  No adjustment 1.0 1.12 (0.87–1.45) 1.32 (1.03–1.69) 0.09 1.12 (1.01–1.24) 0.03 
  Multivariable model I 1.0 1.04 (0.79–1.36) 1.16 (0.89–1.52) 0.52 1.07 (0.97–1.20) 0.19 
  Multivariable model II 1.0 1.04 (0.78–1.37) 1.12 (0.85–1.47) 0.72 1.06 (0.95–1.18) 0.31 
Haemorrhagic stroke (n=486) 122 (25.1%) 172 (35.4%) 192 (39.5%)    
 OR (95% CI)       
  No adjustment 1.0 1.52 (1.17–1.98) 1.63 (1.25–2.12) 0.001 1.21 (1.09–1.34) 0.001 
  Multivariable model I 1.0 1.38 (1.01–1.89) 1.49 (1.09–2.04) 0.03 1.14 (1.01–1.29) 0.03 
  Multivariable model II 1.0 1.32 (0.96–1.82) 1.48 (1.08–2.02) 0.05 1.14 (1.01–1.28) 0.04 

The prognostic value of telomeres in stroke recurrence and post-stroke death was studied by prospectively following up the recruited stroke patients (Figure 1). After adjustment for vascular risk factors, shorter telomere length was not related to the risk of stroke recurrence in the cases with atherothrombotic stroke, lacunar infarction or haemorrhagic stroke. Post-stroke death analysis showed that atherothrombotic stroke patients with shorter telomeres had 69% increased risk [RR (relative risk), 1.69 (95% CI, 1.07–2.67); P=0.02] when comparing the lowest with the highest tertile, but no associations were observed in the follow-up patients with lacunar infarction or hemorrhagic stroke. As antihypertensive therapy might affect the association, we did an additional analysis stratified by antihypertensive therapy and the results were not substantially altered.

RRs (95% CI) of stroke recurrence and post-stroke death by tertiles of telomere length in the follow-up stroke cohort

Figure 1
RRs (95% CI) of stroke recurrence and post-stroke death by tertiles of telomere length in the follow-up stroke cohort

RRs (95% CI) were obtained with multivariate Cox regression models by adjusting for age, gender, and conventional risk factors, including BMI, systolic and diastolic BP, fasting glucose, triacylglycerols, total cholesterol, HDL-cholesterol and LDL (low-density lipoprotein)-cholesterol, smoking status, alcohol intake, diabetes, history of hypertension, previous coronary heart disease, family history of stroke, neurological deficit and medication. The cut-off values of tertile of leucocyte telomere length were in Table 3, with the highest tertile as the reference.

Figure 1
RRs (95% CI) of stroke recurrence and post-stroke death by tertiles of telomere length in the follow-up stroke cohort

RRs (95% CI) were obtained with multivariate Cox regression models by adjusting for age, gender, and conventional risk factors, including BMI, systolic and diastolic BP, fasting glucose, triacylglycerols, total cholesterol, HDL-cholesterol and LDL (low-density lipoprotein)-cholesterol, smoking status, alcohol intake, diabetes, history of hypertension, previous coronary heart disease, family history of stroke, neurological deficit and medication. The cut-off values of tertile of leucocyte telomere length were in Table 3, with the highest tertile as the reference.

Combined effects of telomere length shortening and family history of stroke

We also tested for effect modification by age, gender, history of hypertension, family history of stroke and other vascular risk factors by performing analyses stratified by these variables and by evaluating interaction terms. Decrease of leucocyte telomere length remained to be associated with presence of atherothrombotic stroke and haemorrhagic stroke in various groups (see Supplementary Table S3 at http://www.clinsci.org/cs/125/cs1250027add.htm). As for age, individuals in the age groups of 50–59 years and 60–69 years had an increased presence of stroke per 1−S.D. decrease in telomere length, but those persons in the younger (<50 years) and older (>70 years) groups did not. Magnitude of associations were stronger in persons with family history of stroke than those without, and tests for interaction were statistically significant (Pinteraction=0.05 for atherothrombotic stroke and Pinteraction<0.001 for haemorrhagic stroke).

Next, we further assessed whether telomere length could add additional information to the presence of stroke and the risk of post-stroke death for individuals with family history of stroke. After adjustment for conventional vascular risk factors, we found that those individuals at the lowest tertile of telomeres and also with a family history of stroke had a 2.55-fold (95% CI, 1.87–3.48; Ptrend<0.0001) increased presence of atherothrombotic stroke and a 2.33-fold (95% CI, 1.62–3.36; Ptrend<0.0001) increased presence of haemorrhagic stroke, when compared with subjects at the highest tertile of telomeres and without a family history of stroke (Figure 2). We did not observe combined effects of shorter telomere length and family history of stroke on the risk of post-stroke death in the follow-up stroke cohort.

Presence of atherothrombotic and haemorrhagic stroke according to shorter telomere length and stroke family history

Figure 2
Presence of atherothrombotic and haemorrhagic stroke according to shorter telomere length and stroke family history

The cut-off values of tertile of leucocyte telomere length were defined in Table 3. The common reference category for comparisons was subjects at the highest tertile of telomeres and no family history of stroke. (A) Presence of atherothrombotic stroke, with multivariate logistic regression analysis by adjusting for the same covariates as in Figure 1. (B) Presence of haemorrhagic stroke, with multivariate logistic regression analysis by adjusting for the same covariates as in Figure 1.

Figure 2
Presence of atherothrombotic and haemorrhagic stroke according to shorter telomere length and stroke family history

The cut-off values of tertile of leucocyte telomere length were defined in Table 3. The common reference category for comparisons was subjects at the highest tertile of telomeres and no family history of stroke. (A) Presence of atherothrombotic stroke, with multivariate logistic regression analysis by adjusting for the same covariates as in Figure 1. (B) Presence of haemorrhagic stroke, with multivariate logistic regression analysis by adjusting for the same covariates as in Figure 1.

Correlation between leucocyte telomere length and vascular telomere length in human atherosclerotic plaque lesions

To evaluate whether leucocyte telomere length is an appropriate index of vascular telomere length in diseased vessel wall, we measured telomere length in atherosclerotic carotid plaques and in corresponding circulating leucocytes from patients with carotid atherosclerosis undergoing carotid endarterectomy (n=12). The utilization of human vascular tissues was approved by the ethics committee of Fu Wai Hospital. Clinical characteristics of subjects are shown in Supplementary Table S4 (at http://www.clinsci.org/cs/125/cs1250027add.htm), with means (S.D.) age of 67.4 (9.1) years. Our data showed that the mean telomere length was significantly reduced in blood leucocytes compared with diseased vessel wall specimens of human carotid plaques [median, 0.46 (interquartile range, 0.34–0.54) against 0.56 (interquartile range, 0.49–0.63) respectively; P=0.038]. Adjusting for age and gender, there was a positive correlation between the vessel wall tissue and leucocyte telomere length (partial correlation coefficient γ=0.87; P=0.0001; Supplementary Figure S2 at http://www.clinsci.org/cs/125/cs1250027add.htm).

DISCUSSION

In this large Chinese Stroke Study, we confirmed the association between the shorter leucocyte telomere length and high presence of atherothrombotic stroke, but not of lacunar stroke, and provided novel evidence that telomere attrition contributes to the presence of haemorrhagic stroke, independent of traditional vascular risk factors. Our findings also showed that atherothrombotic stroke patients with shorter telomeres had worse prognosis of post-stroke death during the long-term follow-up. Further assays in human carotid atherosclerotic plaques demonstrated that blood leucocyte telomere length is closely correlated with vascular telomere length in diseased vessel wall.

In addition, we found that those individuals with shorter telomere length and a family history of stroke had a significant 2.55-fold increased presence of atherothrombotic stroke and a 2.33-fold increased presence of haemorrhagic stroke, which might help to screen persons at high risk.

The strengths of the study are in two aspects. First, the large number of stroke patients allowed us to examine the impact of telomeres in different stroke subtypes. In previous studies in Chinese subjects [7,13], the number of haemorrhagic strokes was too small to allow for any conclusion as to whether shorter telomeres have any effect on this particular subtype of stroke. The present study provided evidence that leucocyte telomere attrition is associated with the increased susceptibility to atherothrombotic or haemorrhagic stroke, but not with lacunar infarction. Haemorrhagic stroke results from the spontaneous rupture and bleeding of vessels due to long-term atherosclerosis and arterial stiffness, including microaneurysms due to hypertension, cerebral amyloid angiopathy in the elderly or unidentified risk factors [14]. As for lacunar infarction, lacunes are caused by occlusion of a small deep penetrating artery (with a diameter of <15 mm) in the brain, mainly due to the pathology of lipohyalinosis, fibrinoid degeneration and microatheroma in small vessels. These findings together indicate a key and complex role of telomere attrition in vascular biology. More evidence will be helpful for clarifying the molecular signal pathways by which telomere attrition affects vascular risk.

The second strength of the study included its prospective design, high follow-up rate (95.3%) and long-term follow-up period. This is the largest study investigating whether telomere attrition is associated with the outcome after stroke in patients with atherothrombosis or intracerebral haemorrhage, and our findings showed that shorter telomere is a potential prognostic marker of post-stroke death, particularly in persons with atherothrombosis. Stroke is a major disorder in China [15], even though short telomeres confer a modest RR, the population-attributable fraction may be high.

In the current study, on average, the mean leucocyte telomere length in individuals with stroke was comparable with control subjects chronologically 6 years older, which is in agreement with previous reports [7,16]. Our findings support the hypothesis that those individuals with premature biological aging are at increased risk of stroke development and worse prognosis in comparison with their peers. The question of whether telomere attrition is a causal factor in the pathogenesis of stroke, or whether both telomere attrition and stroke are caused by an unknown causal factor, warrants further investigation. However, emerging evidence support that shorter inherited telomeres might be an important causal contribution to the inherited cardiovascular risk difference in individuals [17,18]. The offspring study showed that shorter telomeres occurred in young, healthy subjects with a family history of coronary artery disease [19,20]. A recent elderly Danish twins study clearly shows that the co-twins with the shorter telomeres are more likely to die first [21]. Leucocyte telomere length is defined at birth [22], and telomeres attrition reflects increased leucocyte turnover as the burden of chronic inflammation and oxidative stress accumulate during the individual's life course, which is believed to be a common basis in the progression of atherosclerosis and age-dependent vascular diseases [8]. Moreover, our additional assays, using paired samples of blood leucocyte and human carotid atherosclerotic plaque lesions in the same subject, suggest that blood leucocyte telomeres may be used as an effective surrogate for vascular aging. Given the small sample size, longitudinal assessment of telomere dynamics in large cohorts and clinic trials based on this factor should further test and confirm its role in the prevention and treatment of vascular diseases.

Our study had some potential limitations. First, since the present study is a case-control study and prospective cohort study for recruited stroke patients, selection bias of the control and stroke populations cannot be excluded. In addition, information on statin usage and physical activity was lacking. Recent evidence shows that statin treatment and exercise may prevent yearly loss of telomere length [2325], and also reduce stroke recurrence and may improve prognosis post-stroke [26]. However, based on a national survey on serum cholesterol levels in China, awareness, treatment, and control rates of hypercholesterolaemia in the Chinese population were less than 8.0, 3.4 and 2.0%, respectively [27], it therefore seems unlikely to affect our results meaningfully. Secondly, we measured mean telomere length of leucocytes; however, this can be affected by changes in leucocyte composition. Accelerated telomere shortening in leucocyte subpopulations of patients with coronary heart disease has been found to be highly conserved throughout the haematopoietic system, and particularly significant in cytotoxic T lymphocytes of cytomegalovirus-seropositive patients [28]. Third, the potential effects of genetic variants on telomere length and diseases risk were not determined, whereas several telomere-associated loci have been identified in some studies [29–31]. More replication studies of leucocyte telomere length are needed to confirm the clinical utility as an informative marker of cardiovascular diseases including stroke.

FUNDING

This work was supported by the Ministry of Science and Technology of China [grant number 2011CB503901 (to R.H.)] and by the National Natural Science of China Foundation [grant number81070172 (to W.Z.)].

AUTHOR CONTRIBUTION

Weili Zhang and Rutai Hui contributed to the study conception and design, drafting of the paper, and analysing and interpretating the data. Yu Chen and Yuyao Wang contributed to the analysis of data and critical revision of the paper. Peng Liu, Mei Zhang, Channa Zhang and Frank Hu contributed to the interpretation of data and critical revision of the paper for important intellectual content. All authors read and approved the final submitted paper.

We thank Dr Hu Ding and Professor Daowen Wang (Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology) for their excellent technical support and the collaborators in this Multicenter Stroke Study for data collection.

Abbreviations

     
  • BMI

    body mass index

  •  
  • BP

    blood pressure

  •  
  • CHD

    coronary heart disease

  •  
  • CI

    confidence interval

  •  
  • CT

    computed tomography

  •  
  • MRI

    magnetic resonance imaging

  •  
  • HDL-cholesterol

    high-density lipoprotein-cholesterol

  •  
  • OR

    odds ratio

  •  
  • RR

    relative risk

References

References
1
Calado
R. T.
Young
N. S.
Telomere diseases
N. Engl. J. Med.
2009
, vol. 
361
 (pg. 
2353
-
2365
)
2
Blackburn
E. H.
Structure and function of telomeres
Nature
1991
, vol. 
350
 (pg. 
569
-
573
)
3
Miyauchi
H.
Yoshida
T.
Ishida
Y.
Yoshida
H.
Komuro
I.
Endothelial cell senescence in human atherosclerosis: role of telomere in endothelial dysfunction
Circulation
2002
, vol. 
105
 (pg. 
1541
-
1544
)
4
Zee
R. Y.
Castonguay
A. J.
Barton
N. S.
Ridker
P. M.
Relative leukocyte telomere length and risk of incident ischemic stroke in men: a prospective, nested case-control approach
Rejuvenation Res.
2010
, vol. 
13
 (pg. 
411
-
414
)
5
Fitzpatrick
A. L.
Kronmal
R. A.
Kimura
M.
Gardner
J. P.
Psaty
B. M.
Jenny
N. S.
Tracy
R. P.
Hardikar
S.
Aviv
A.
Leukocyte telomere length and mortality in the cardiovascular health study
J. Gerontol. A. Biol. Sci. Med. Sci.
2011
, vol. 
66
 (pg. 
421
-
429
)
6
Fyhrquist
F.
Silventoinen
K.
Saijonmaa
O.
Kontula
K.
Devereux
R. B.
de Faire
U.
Os
I.
Dahlöf
B.
Telomere length and cardiovascular risk in hypertensive patients with left ventricular hypertrophy: the Life Study
J. Hum. Hypertens.
2011
, vol. 
25
 (pg. 
711
-
718
)
7
Ding
H.
Chen
C.
Shaffer
J. R.
Liu
L.
Xu
Y.
Wang
X.
Hui
R.
Wang
D. W.
Telomere length and risk of stroke in Chinese
Stroke
2012
, vol. 
43
 (pg. 
658
-
663
)
8
Aviv
A.
Telomeres and human somatic fitness
J. Gerontol. A. Biol. Sci. Med. Sci.
2006
, vol. 
61
 (pg. 
871
-
873
)
9
Wilson
W. R.
Herbert
K. E.
Mistry
Y.
Stevens
S. E.
Patel
H. R.
Hastings
R. A.
Thompson
M. M.
Williams
B.
Blood leucocyte telomere DNA content predicts vascular telomere DNA content in humans with and without vascular disease
Eur. Heart J.
2008
, vol. 
29
 (pg. 
2689
-
2694
)
10
Zhang
W.
Sun
K.
Zhen
Y.
Wang
D.
Wang
Y.
Chen
J.
Xu
J.
Hu
F. B.
Hui
R.
VEGF receptor-2 variants are associated with susceptibility to stroke and recurrence
Stroke
2009
, vol. 
40
 (pg. 
2720
-
2726
)
11
Burn
J.
Dennis
M.
Bamford
J.
Sandercock
P.
Wade
D.
Warlow
C.
Long-term risk of recurrent stroke after a first-ever stroke: the oxfordshire community stroke project
Stroke
1994
, vol. 
25
 (pg. 
333
-
337
)
12
Cawthon
R. M.
Telomere measurement by quantitative PCR
Nucleic Acids Res.
2002
, vol. 
30
 pg. 
e 47
 
13
Yang
Z.
Huang
X.
Jiang
H.
Zhang
Y.
Liu
H.
Qin
C.
Eisner
G. M.
Jose
P. A.
Rudolph
L.
Ju
Z.
Short telomeres and prognosis of hypertension in a Chinese population
Hypertension
2009
, vol. 
53
 (pg. 
639
-
645
)
14
Qureshi
A. I.
Tuhrim
S.
Broderick
J. P.
Batjer
H. H.
Hondo
H.
Hanley
D. F.
Spontaneous intracerebral hemorrhage
N. Engl. J. Med.
2001
, vol. 
344
 (pg. 
1450
-
1460
)
15
Reed
D. M.
The paradox of high risk of stroke in populations with low risk of coronary heart disease
Am. J. Epidemiol.
1990
, vol. 
131
 (pg. 
579
-
588
)
16
Brouilette
S. W.
Moore
J. S.
McMahon
A. D.
Thompson
J. R.
Ford
I.
Shepherd
J.
Packard
C. J.
Samani
N. J.
West of Scotland Coronary Prevention Study Group
Telomere length, risk of coronary heart disease, and statin treatment in the West of Scotland Primary Prevention Study: a nested case-control study
Lancet
2007
, vol. 
369
 (pg. 
107
-
114
)
17
Matthews
C.
Gorenne
I.
Scott
S.
Figg
N.
Kirkpatrick
P.
Ritchie
A.
Goddard
M.
Bennett
M.
Vascular smooth muscle cells undergo telomere-based senescence in human atherosclerosis: effects of telomerase and oxidative stress
Circ. Res.
2006
, vol. 
99
 (pg. 
156
-
164
)
18
Chang
E.
Harley
C. B.
Telomere length and replicative aging in human vascular tissues
Proc. Natl. Acad. Sci. U.S.A.
1995
, vol. 
92
 (pg. 
11190
-
11194
)
19
Brouilette
S. W.
Whittaker
A.
Stevens
S. E.
van der Harst
P.
Goodall
A. H.
Samani
N. J.
Telomere length is shorter in healthy offspring of subjects with coronary artery disease: support for the telomere hypothesis
Heart
2008
, vol. 
94
 (pg. 
422
-
425
)
20
Farzaneh-Far
R.
Cawthon
R. M.
Na
B.
Browner
W. S.
Schiller
N. B.
Whooley
M. A.
Prognostic value of leukocyte telomere length in patients with stable coronary artery disease: data from the Heart and Soul Study
Arterioscler. Thromb. Vasc. Biol.
2008
, vol. 
28
 (pg. 
1379
-
1384
)
21
Kimura
M.
Hjelmborg
J. V.
Gardner
J. P.
Bathum
L.
Brimacombe
M.
Lu
X.
Christiansen
L.
Vaupel
J. W.
Aviv
A.
Christensen
K.
Short leukocyte telomeres forecast mortality: a study in elderly Danish twins
Am. J. Epidemiol.
2008
, vol. 
167
 (pg. 
799
-
806
)
22
Okuda
K.
Bardeguez
A.
Gardner
J. P.
Rodriguez
P.
Ganesh
V.
Kimura
M.
Skurnick
J.
Awad
G.
Aviv
A.
Telomere length in the newborn
Pediatr. Res.
2002
, vol. 
52
 (pg. 
377
-
381
)
23
Spyridopoulos
I.
Haendeler
J.
Urbich
C.
Brummendorf
T. H.
Oh
H.
Schneider
M. D.
Zeiher
A. M.
Dimmeler
S.
Statins enhance migratory capacity by upregulation of the telomere repeat-binding factor TRF2 in endothelial progenitor cells
Circulation
2004
, vol. 
110
 (pg. 
3136
-
3142
)
24
Cherkas
L. F.
Hunkin
J. L.
Kato
B. S.
Richards
J. B.
Gardner
J. P.
Surdulescu
G. L.
Kimura
M.
Lu
X.
Spector
T. D.
Aviv
A.
The association between physical activity in leisure time and leukocyte telomere lengthArch. Int
Med.
2008
, vol. 
168
 (pg. 
154
-
158
)
25
Werner
C.
Hanhoun
M.
Widmann
T.
Kazakov
A.
Semenov
A.
Pöss
J.
Bauersachs
J.
Thum
T.
Pfreundschuh
M.
Müller
P.
, et al. 
Effects of physical exercise on myocardial telomere-regulating proteins, survival pathways, and apoptosis
J. Am. Coll. Cardiol.
2008
, vol. 
52
 (pg. 
470
-
482
)
26
Nassief
A.
Marsh
J. D.
Statin therapy for stroke prevention
Stroke
2008
, vol. 
39
 (pg. 
1042
-
1048
)
27
He
J.
Gu
D.
Reynolds
K.
Wu
X.
Muntner
P.
Zhao
J.
Chen
J.
Liu
D.
Mo
J.
Whelton
P. K.
InterASIA Collaborative Group
Serum total and lipoprotein cholesterol levels and awareness, treatment, and control of hypercholesterolemia in China
Circulation
2004
, vol. 
110
 (pg. 
405
-
411
)
28
Spyridopoulos
I.
Hoffmann
J.
Aicher
A.
Brümmendorf
T. H.
Doerr
H. W.
Zeiher
A. M.
Dimmeler
S.
Accelerated telomere shortening in leukocyte subpopulations of patients with coronary heart disease: role of cytomegalovirus seropositivity
Circulation
2009
, vol. 
120
 (pg. 
1364
-
1372
)
29
Andrew
T.
Aviv
A.
Falchi
M.
Surdulescu
G. L.
Gardner
J. P.
Lu
X.
Kimura
M.
Kato
B. S.
Valdes
A. M.
Spector
T. D.
Mapping genetic loci that determine leukocyte telomere length in a large sample of unselected female sibling pairs
Am. J. Hum. Genet.
2006
, vol. 
78
 (pg. 
480
-
486
)
30
Codd
V.
Mangino
M.
van der Harst
P.
Braund
P. S.
Kaiser
M.
Beveridge
A. J.
Rafelt
S.
Moore
J.
Nelson
C.
Soranzo
N.
Common variants near terc are associated with mean telomere length
Nat. Genet.
2010
, vol. 
42
 (pg. 
197
-
199
)
31
Matsubara
Y.
Murata
M.
Watanabe
K.
Saito
I.
Miyaki
K.
Omae
K.
Ishikawa
M.
Matsushita
K.
Iwanaga
S.
Ogawa
S.
, et al. 
Coronary artery disease and a functional polymorphism of hTERT
Biochem. Biophys. Res. Commun.
2006
, vol. 
348
 (pg. 
669
-
672
)

Supplementary data