Abstract
Background
Hepatocyte growth factor (HGF) is a potent angiogenic factor and may play a role in the development and progression of atherosclerotic lesions, the underlying mechanism of cardiovascular disease. However, there have been no prospective studies examining the relationship between HGF levels and risk of stroke.
Methods and Results
We conducted a nested case-control study (972 incident stroke cases and 1:1 age- and race-matched controls) to prospectively evaluate the association between plasma HGF and risk of ischemic stroke within the Women’s Health Initiative Observational Study, a cohort of postmenopausal women aged 50–79 years. Baseline HGF levels were correlated positively with body mass index (BMI), systolic blood pressure, low-density lipoprotein cholesterol, insulin resistance, and inflammatory markers such as C-reactive protein, and inversely with high-density lipoprotein cholesterol (all P-values <0.05). Baseline HGF levels were higher among cases than controls (geometric means 601.8 vs. 523.2 pg/mL, p = 0.003). Furthermore, the risk of incident ischemic stroke was significantly greater amongst women in the highest versus lowest quartile of plasma HGF levels (odds ratio [OR] = 1.46; 95% confidence interval [CI]: 1.12–1.91; Ptrend = 0.003), in a conditional logistic regression model that adjusted for BMI. These results were only slightly attenuated after further adjustment for additional stroke risk factors (OR=1.39; 95% CI=1.04–1.85, Ptrend=0.023).
Conclusions
Circulating levels of HGF are associated with an increased risk of incident ischemic stroke, independent of obesity and other risk factors for cardiovascular disease among postmenopausal women aged 50–79 years.
Keywords: Hepatocyte growth factor, ischemic stroke, women
Hepatocyte growth factor (HGF), initially described as a mitogen for hepatocytes,1–2 is a potent angiogenic factor and endothelium-specific growth factor that affects a wide range of tissues. Its receptor, a transmembrane tyrosine kinase encoded by the proto-oncogene c-Met, is expressed by a variety of cell types (e.g., epithelial cells, hepatocytes, and vascular endothelial cells).3–4 Activation of this receptor in vascular endothelial cells leads to cell dissociation and migration, proliferation, protease production, invasion, and neovascularization.4 Through these functions, it is hypothesized that HGF may play a role in the natural history of atherosclerosis and hence the pathogenesis of cardiovascular disease (CVD), including ischemic stroke.4–7
Consistent with this view, recent data in humans have demonstrated expression of c-Met and HGF in atherosclerotic plaques.8 Interestingly, serum levels of HGF correlate positively with several CVD risk factors, including obesity,9 hypertension,10 and insulin resistance,11 and inversely with high-density lipoprotein (HDL)-cholesterol.11 Furthermore, cross-sectional data have suggested that serum HGF levels are higher in patients with acute ischemic stroke than in healthy controls.12–13 To the best of our knowledge, however, there have been no prospective studies that have evaluated this association. Thus, it is unclear whether (i) high HGF levels are a risk factor for ischemic stroke or a result of it, and (ii) HGF is an independent risk factor for stroke, or merely a marker for other CVD risk factors. To address these issues, we conducted a nested case-control study within the Women’s Health Initiative Observational Study (WHI-OS) and evaluated the association between circulating HGF levels and risk of developing ischemic stroke.
METHODS
Study Population
The WHI-OS is an ongoing, ethnically and geographically diverse prospective study, with long-term follow-up of 93,676 participants at 40 clinical centers in the United States, to examine the risk factors for subsequent development of several health outcomes.14–16 Briefly, postmenopausal women aged 50 to 79 years were enrolled between 1993 to 1998. At baseline, women were queried about lifestyle factors, medical history, and personal habits. A physical examination was performed to obtain height, weight, and blood pressure. Fasting blood samples were collected, centrifuged, frozen on site at –70°C, and later shipped to a central specimen repository. As of September 2005, 4.7% of the original 93,676 WHI-OS subjects had been lost-to-follow-up.
The Hormones and Biomarkers Predicting Stroke study (HaBPS)
The HaBPS was a nested case-control study based in the WHI-OS. The goal was to determine if baseline circulating levels of various biomarkers, including adipokines, novel lipoproteins, inflammatory cytokines, and hemostatis markers, were associated with risk of ischemic stroke.17–18 Women were ineligible if they had a history of myocardial infarction or stroke at baseline.
Medical records were obtained for all potential stroke events that were identified through self-report at annual contacts; adjudication was performed locally by trained physicians who assigned a diagnosis according to standardized criteria. Only stroke events that required hospitalization were considered as potential outcomes and transient ischemic attacks were not included. All locally adjudicated stroke events were then sent for central adjudication by a team of study neurologists. Strokes were classified as ischemic or hemorrhagic on review of reports of brain imaging studies. Ischemic stroke was defined as the rapid onset of a persistent neurologic deficit attributed to an obstruction lasting >24 hours without evidence of other causes, unless death supervened or there was a demonstrable lesion compatible with acute stroke on computed tomography or MRI scan. The subtypes of ischemic stroke were classified using Trial of Org 10172 in Acute Stroke Treatment (TOAST) criteria19 as large artery atherothrombosis, cardioembolic, lacunar (small vessel), or undetermined etiology.
A total of 972 incident ischemic stroke cases were identified from study baseline to July 1, 2003. One control was selected from the risk set at the time of the case diagnosis and matched to each case on age at baseline (±2 years), race/ethnicity, date of study enrollment (±3 months), and follow-up time.
Plasma HGF levels were measured by a multiplex assay (Human Adipokine Panel B, Millipore, Billerica, MA).20 The inter-assay coefficient of variation for HGF assay was 11.7%. We previously reported the 3-year intra-individual correlation coefficient of HGF in a separate study population to be 0.91 (95% CI = 0.86 – 0.97),21 indicating that circulating levels of HGF are relatively stable, and that a single measurement at baseline may reflect an individual’s long term exposure to this growth factor.
Statistical Analysis
We first assessed the associations of HGF levels with reported risk factors for ischemic stroke (e.g., obesity, smoking, and hypertension) among controls. These baseline characteristics were compared by quartiles of HGF using ANOVA and chi-square test. Cases and controls were then compared with respect to the reported risk factors for ischemic stroke at baseline; because of the matched case-control design, univariable conditional logistic regression and paired t-test were used.
Multivariable conditional logistic regression was used to evaluate if HGF levels were associated with risk of incident ischemic stroke, while adjusting for confounding variables identified from the two analyses described above. HGF was modeled in quartiles to avoid assuming linearity. Tests of linear trend across quartiles were conducted by assigning the median value for each category, fitting this as a continuous variable in the model, and assessing statistical significance using Wald test. The multivariable model adjusted for the following baseline covariates that were associated with ischemic stroke in this study population: BMI (continuous), physical activity [metabolic equivalent tasks (METs) per week in quartiles], smoking, use of non-steroidal anti-inflammatory drugs (NSAIDs), use of hypertension medication, systolic blood pressure (continuous), diabetes (self-reported treatment or a fasting glucose level ≥126 mg/dL), and history of coronary or artery disease (none, 1, or ≥ 2 of the following conditions: angina, congestive heart failure, coronary revascularization procedure, atrial fibrillation, or peripheral artery disease). Among all the biomarkers examined in HaBPS, C-reactive protein (CRP) was the most robustly associated with the risk of ischemic stroke,17 so we also evaluated the impact of additional adjustment for CRP on the HGF association with stroke. Certain risk factors were associated with ischemic stroke in previous studies (e.g., triglycerides, LDL-cholesterol level, and postmenopausal hormone therapy) or in our univariable analyses [e.g., interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α)], but they were no longer significant after adjusting for the aforementioned covariates, and they were not included in the regression model. Finally, to evaluate if the association of HGF and ischemic stroke could be modified by other stroke risk factors (e.g., BMI, hypertension, and CRP), interaction terms were included one at a time into the final multivariable model and their significance was determined by Wald test.
All statistical analyses were performed using SAS® software version 9.1 (SAS Institute, Cary, NC), and p-values less than 0.05 were considered statistically significant.
RESULTS
Circulating HGF levels were positively associated with CVD risk factors such as age, BMI, waist circumference, diabetes, systolic blood pressure, total and LDL-cholesterol, and HOMA-IR (homeostatic assessment for insulin resistance), and inversely associated with HDL-cholesterol levels (Table 1). In addition, relatively higher HGF levels were associated with increased levels of inflammatory markers, including CRP, IL-6, and TNF-α.
Table 1.
Baseline characteristics by quartiles of HGF among controls *
| Quartile 1 (n=242) (Median=229.8 pg/mL) |
Quartile 2 (n=242) (Median=500.0 pg/mL) |
Quartile 3 (n=242) (Median=758.3 pg/mL) |
Quartile 4 (n=241) (Median=1231.7 pg/mL) |
P value† |
|
|---|---|---|---|---|---|
| Sociodemographic factors | |||||
| Age, years, mean (SD) | 67.6 (6.3) | 68.3 (6.3) | 69.1 (6.5) | 69.9 (6.2) | 0.001 |
| Race-ethnicity, n (%) | 0.181 | ||||
| White Non-Hispanic | 207 (86) | 209 (86) | 212 (88) | 201 (83) | |
| Black | 26 (11) | 14 (6) | 17 (7) | 23 (10) | |
| Other | 9 (4) | 19 (8) | 13 (5) | 17 (7) | |
| Education, n (%) | 0.029‡ | ||||
| ≤ High school | 48 (20) | 59 (24) | 51 (21) | 67 (28) | |
| Some college/post secondary training | 92 (38) | 82 (34) | 92 (38) | 94 (39) | |
| College graduate or higher | 102 (42) | 100 (41) | 97 (40) | 80 (33) | |
| Income, n (%) | 0.027‡ | ||||
| <$35,000 | 95 (45) | 103 (46) | 109 (48) | 123 (55) | |
| $35,000–$49,999 | 37 (18) | 54 (24) | 36 (16) | 45 (20) | |
| ≥$50,000 | 78 (37) | 66 (30) | 84 (37) | 56 (25) | |
| Lifestyle factors | |||||
| Hormone therapy, n (%) | 0.006 | ||||
| Never | 70 (29) | 82 (34) | 76 (32) | 94 (40) | |
| Past | 48 (20) | 53 (22) | 71 (30) | 60 (26) | |
| Current | 120 (50) | 104 (44) | 90 (38) | 80 (34) | |
| Smoking Status, n (%) | 0.792 | ||||
| Never | 137 (57) | 132 (55) | 125 (52) | 131 (55) | |
| Former | 93 (39) | 95 (40) | 107 (44) | 101 (42) | |
| Current | 9 (4) | 12 (5) | 9 (4) | 7 (3) | |
| Physical Activity, METs/week, mean (SD) | 15.0 (14.0) | 15.3 (14.9) | 13.8 (14.9) | 12.8 (12.5) | 0.177 |
| Anthropometric factors | |||||
| BMI, kg/m2, mean (SD) | 26.5 (5.3) | 26.3 (4.7) | 26.8 (5.4) | 28.4 (5.6) | < 0.0001 |
| Waist Circumference, cm, mean (SD) | 82.8 (11.2) | 82.8 (11.3) | 84.2 (12.4) | 89.1 (13.0) | < 0.0001 |
| Medical History and examination | |||||
| Current NSAID use, n (%) | 84 (35) | 77 (32) | 82 (34) | 88 (37) | 0.748 |
| Diabetes, n (%) | 13 (5) | 15 (6) | 21 (9) | 31 (13) | 0.014 |
| History of coronary and arterial disease, n (%) | 32 (13) | 25 (10) | 29 (12) | 40 (17) | 0.212 |
| Current hypertension medication use, n (%) | 72 (30) | 74 (31) | 88 (36) | 103 (43) | 0.009 |
| Systolic blood pressure, mmHg, mean (SD) | 128.8 (19.3) | 127.9 (16.8) | 130.9 (17.5) | 132.8 (18.3) | 0.012 |
| Diastolic blood pressure, mmHg, mean (SD) | 73.8 (9.2) | 73.2 (9.2) | 74.9 (9.1) | 74.4 (10.6) | 0.254 |
| Laboratory measurements | |||||
| Total Cholesterol, mg/dL, mean (SD) | 229.7 (39.2) | 234.5 (38.0) | 234.2 (38.1) | 224.6 (37.6) | 0.013 |
| HDL-cholesterol, mg/dL, mean (SD) | 63.5 (17.3) | 60.9 (16.3) | 60.1 (16.3) | 54.8 (14.5) | < 0.0001 |
| LDL-cholesterol, mg/dL, mean (SD) | 134.5 (38.6) | 142.6 (37.5) | 143.1 (36.5) | 135.5 (33.3) | 0.011 |
| Triglycerides, mg/dL, mean (SD) | 156.4 (72.3) | 155.0 (69.4) | 156.1 (79.2) | 175.8 (96.4) | 0.010 |
| Ln HOMA-IR, mean (SD) ¶ | 0.25 (0.74) | 0.32 (0.67) | 0.40 (0.80) | 0.71 (0.76) | < 0.0001 |
| Ln CRP, mg/L, mean (SD) | 0.79 (1.10) | 0.88 (1.05) | 0.77 (1.08) | 1.11 (1.12) | 0.002 |
| Ln IL-6, pg/mL, mean (SD) | 0.60 (0.64) | 0.63 (0.61) | 0.63 (0.58) | 0.84 (0.68) | < 0.0001 |
| Ln TNF-α, pg/mL, mean (SD) | 0.29 (0.44) | 0.39 (0.48) | 0.35 (0.48) | 0.48 (0.64) | 0.001 |
The n presented may not add to the total due to missing information. Abbreviations used: BMI: body mass index; CRP: C reactive protein; HDL: high density lipoprotein; HOMA-IR: homeostatic model assessment for insulin resistance; IL-6: interleukin 6; LDL: low density lipoprotein; METS: metabolic equivalent tasks; NSAID: Non-steroidal anti-inflammatory drugs; SD: standard deviation; TNF-α: tumor necrosis factor-alpha.
P value was obtained by chi-square test for categorical variables and ANOVA for continuous variables, unless otherwise specified.
P value from Mantel-Haenszel chi-square test for trend.
HOMA-IR was calculated as [Insulin (mU/ml)×glucose (mmol/L)] /22.5.
Table 2 shows the univariable comparisons between cases and controls. As expected, ischemic stroke was associated positively with many established CVD risk factors, namely current smoking, BMI, waist circumference, diabetes, history of coronary and arterial disease, systolic blood pressure, and HOMA-IR, and inversely with physical activity and HDL-cholesterol levels. Furthermore, cases had higher levels of inflammatory markers than controls. Geometric mean levels of HGF were 601.8 pg/mL in cases and 523.2 pg/mL in controls (p=0.003).
Table 2.
Baseline characteristics among cases of ischemic stroke and matched controls
| Controls (n = 972) |
Cases (n = 972) |
P value† | |
|---|---|---|---|
| Sociodemographic factors | |||
| Age, years, mean (SD) | 68.7 (6.4) | 68.7 (6.4) | NA |
| Race-ethnicity, n (%) | NA | ||
| White Non-Hispanic | 833 (86) | 833 (86) | |
| Black | 80 (8) | 80 (8) | |
| Other | 59 (6) | 59 (6) | |
| Education, n (%) | |||
| ≤ High school | 226 (23) | 237 (25) | Reference |
| Some college/post secondary training | 362 (37) | 394 (41) | 0.741 |
| College graduate or higher | 381 (39) | 333 (35) | 0.115 |
| Income, n (%) | |||
| <$35,000 | 432 (48) | 482 (53) | Reference |
| $35,000–$49,999 | 172 (19) | 173 (19) | 0.223 |
| ≥$50,000 | 287 (32) | 247 (27) | 0.031 |
| Lifestyle factors | |||
| Hormone therapy, n (%) | |||
| Never | 323 (34) | 309 (32) | Reference |
| Past | 235 (25) | 226 (24) | 0.902 |
| Current | 395 (41) | 416 (44) | 0.306 |
| Smoking Status, n (%) | |||
| Never | 526 (55) | 505 (53) | Reference |
| Former | 400 (42) | 377 (39) | 0.728 |
| Current | 37 (4) | 79 (8) | 0.0003 |
| Physical Activity, METs/week, mean (SD) | 14.2 (14.1) | 12.4 (13.8) | 0.004 |
| Anthropometric factors | |||
| BMI, kg/m2, mean (SD) | 27.0 (5.3) | 27.7 (5.9) | 0.007 |
| Waist Circumference, cm, mean (SD) | 84.7 (12.3) | 87.3 (13.6) | < 0.0001 |
| Medical History and examination | |||
| Current NSAID use, n (%) | 332 (34) | 406 (42) | 0.001 |
| Diabetes, n (%) | 81 (8) | 162 (17) | < 0.0001 |
| History of coronary & arterial disease, n (%) | 126 (13) | 201 (21) | < 0.0001 |
| Current hypertension medication use, n (%) | 340 (35) | 461 (47) | < 0.0001 |
| Systolic blood pressure, mmHg, mean (SD) | 130.1 (18.0) | 137.2 (19.4) | < 0.0001 |
| Diastolic blood pressure, mmHg, mean (SD) | 74.1 (9.5) | 75.5 (10.1) | 0.002 |
| Laboratory measurements | |||
| Total Cholesterol, mg/dL, mean (SD) | 230.8 (38.3) | 233.2 (39.2) | 0.154 |
| HDL-cholesterol, mg/dL, mean (SD) | 59.8 (16.4) | 57.2 (16.2) | 0.0004 |
| LDL-cholesterol, mg/dL, mean (SD) | 139.0 (36.7) | 140.8 (37.4) | 0.313 |
| Triglycerides, mg/dL, mean (SD) | 160.9 (80.4) | 179.1 (89.9) | < 0.0001 |
| Ln HOMA-IR, mean (SD) | 0.42 (0.76) | 0.62 (0.84) | < 0.0001 |
| Ln CRP, mg/L, mean (SD) | 0.88 (1.10) | 1.20 (1.10) | < 0.0001 |
| Ln IL-6, pg/mL, mean (SD) | 0.68 (0.64) | 0.88 (0.69) | < 0.0001 |
| Ln TNF-α, pg/mL, mean (SD) | 0.38 (0.52) | 0.42 (0.56) | 0.039 |
| Ln HGF, pg/mL, mean (SD) | 6.26 (1.09) | 6.40 (0.96) | 0.003 |
The n presented may not add to the total due to missing information.
P value was obtained by conditional logistic regression for categorical variables and paired t-test for continuous variables; NA = matched variable, statistics not performed.
Table 3 shows the association between quartiles of HGF levels and risk of incident stroke. In the conditional logistic regression model accounting for only the matching factors (age and race/ethnicity), the odds ratios (ORs) for incident stroke with increasing quartiles of HGF were 1.0, 1.13, 1.38 and 1.49 (p for trend: 0.001). Since HGF is an adipokine secreted by adipocytes, we added BMI into the model to examine whether HGF was merely a marker for adiposity or its effect on risk of stroke was independent of adiposity. The HGF-stroke association was only slightly attenuated when BMI was adjusted for in the model, with the OR comparing the fourth to the first quartile of HGF being 1.46 (95% CI: 1.12, 1.91; p for trend = 0.003). This association remained significant even after additional adjustment for established risk factors for stroke; the OR between extreme quartiles of HGF was 1.39 (95% CI: 1.04, 1.85; p for trend = 0.023). Further adjustment for CRP did not impact the HGF results; the OR for ischemic stroke comparing the fourth to the first quartile was 1.37 (95% CI: 1.02, 1.83; p for trend: 0.026). In addition, we did not observe any effect modification of the HGF-stroke association by the established stroke risk factors, and the results did not vary by subtype of ischemic stroke (data not shown).
Table 3.
Odds ratios (95% CI) from conditional logistic regression analysis for the association of HGF with incident ischemic stroke
| Quartiles | P for trend |
||||
|---|---|---|---|---|---|
| Q1 | Q2 | Q3 | Q4 | ||
| Number of cases | 194 | 221 | 269 | 281 | |
| Model 1* | Reference | 1.13 (0.87–1.46) | 1.38 (1.07–1.79) | 1.49 (1.14–1.93) | 0.001 |
| Model 2† | Reference | 1.15 (0.88–1.50) | 1.42 (1.09–1.84) | 1.46 (1.12–1.91) | 0.003 |
| Model 3‡ | Reference | 1.16 (0.87–1.54) | 1.32 (0.99–1.74) | 1.39 (1.04–1.85) | 0.023 |
Model 1 = crude (conditional on matching factors - age and race/ethnicity)
Model 2 = Model 1 + BMI
Model 3 = Model 2 + current smoking, physical activity, NSAIDs use, hypertension medication use, systolic blood pressure, history of cardiovascular disease, and diabetes
DISCUSSION
In this prospective study, we found circulating HGF levels to be positively associated with risk of ischemic stroke among postmenopausal women after accounting for established risk factors. HGF may increase the risk of ischemic stroke by promoting progression of atherosclerotic lesions, a critical step in the development of CVD, through its function as a potent angiogenic factor and its role in stimulating migration of vascular endothelial cells. High HGF levels, either in systemic circulation or locally induced by LDL-cholesterol within atherosclerotic plaques, may lead to plaque neovascularization.8, 22 Neovascularization, in turn, facilitates infiltration of leukocytes and inflammatory stimuli, which further enhance angiogenesis in atherosclerotic lesions.23–24 The resulting increase in microvascular density and chronic inflammation may contribute to plaque instability, such as intraplaque haemorrhage and plaque rupture23, which then lead to an acute ischemic stroke.
Previous studies provided initial data linking HGF with atherosclerosis and ischemic stroke. HGF protein was detected within carotid atherosclerotic lesions but not in normal vessels, 25 and circulating HGF levels were correlated with severity of carotid atherosclerosis measured by intima medial thickness.7 Further, a single nucleotide polymorphism in the intronic region of the HGF gene (T43839A) was associated with carotid atherosclerosis in women.26 Two other studies found that circulating HGF was elevated among cases with acute ischemic stroke compared to controls.12–13 However, all these prior data were cross-sectional studies with small sample sizes, involving fewer than 30 patients. Our nested case-control study provided prospective data supporting the role of HGF in pathogenesis of ischemic stroke.
Our study and most previous studies found that HGF was associated with several CVD risk factors. As an adipokine expressed by white adipose tissue,27–29 HGF levels are positively correlated with BMI in several studies.9, 11, 30–32 In addition, HGF levels have been observed to correlate positively with levels of insulin, insulin resistance,11, 31 blood pressure, and triglyceride, but inversely correlated with HDL-cholesterol.11, 31–32 High levels of HGF have also been associated with increased severity of hypertension,33–35 type 2 diabetes,32 and increasing number of abnormalities in the components of the metabolic syndrome.11, 32 In our study, the association between HGF and ischemic stroke remained significant after adjusting for other CVD risk factors in a multivariable regression model; vice versa, after accounting for the effect of HGF, the associations of these CVD risk factors with stroke were unchanged. Therefore, it seems unlikely that circulating levels of HGF are simply a marker for known CVD risk factors. Moreover, many of these adjusted CVD risk factors are related to obesity (e.g., diabetes, CRP, systolic blood pressure, and history of coronary artery diseases), so our multivariable results also support the notions that even though HGF is an adipokine, the mechanism underlying its association with ischemic stroke may go beyond the obesity-metabolic pathway and involve the potent angiogenic bioactivity of HGF.
Nevertheless, the strength of association between HGF and stroke (adjusted OR = 1.39) was moderate in comparison with other established risk factors. For example, in our multivariable regression model, the adjusted ORs were 1.72 for diabetes and 2.33 for smoking (data not shown). It us possible that HGF may have limited value, above and beyond those from the established risk factors, in identifying individuals at risk for stroke.
Our study has a few limitations that warrant discussion. First, HGF might be a marker of other risk factors yet to be identified rather than a causal factor in the pathogenesis of ischemic stroke. Secondly, the relationship of circulating levels of HGF with those within the target tissue is not clear and will require further investigation. Lastly, this study was conducted among postmenopausal women, and our results need to be confirmed in other populations, particularly premenopausal women and men.
In conclusion, our study demonstrated a prospective association between relatively high plasma HGF levels and subsequent development of ischemic stroke among postmenopausal women aged 50–79 years, independent of established stroke risk factors. Future studies, including studies in men and in premenopausal women, are needed to further elucidate the role of HGF in the etiopathogenesis of ischemic stroke.
ACKNOWLEDGEMENT
The research on which this publication is based was funded by contract N01-WH-74310 (to Dr. Ho) with the National Heart, Lung, and Blood Institute. The Hormones and Biomarkers Predicting Stroke (HaBPS) Study was supported by Grant Number R01NS042618 (to Dr.Wassertheil-Smoller) from the National Institutes of Neurological Disorders and Stroke. The Key investigators involved in the HaBPS Study: Albert Einstein College of Medicine: Sylvia Wassertheil-Smoller, Robert Kaplan, Aileen McGinn; Fred Hutchinson Cancer Center: Charles Kooperberg; NIH: John Lynch; State University of New York Downstate Medical Center: Daniel Rosenbaum, Alison E. Baird; Boston University: Philip Wolf. The authors thank Ms. Dan Wang, MSc, for her assistance in data analysis.
The Women's Health Initiative (WHI) program is funded by the National Heart, Lung, and Blood Institute, National Institutes of Health, U.S. Department of Health and Human Services through contracts N01WH22110, 24152, 32100-2, 32105-6, 32108-9, 32111-13, 32115, 32118-32119, 32122, 42107-26, 42129-32, and 44221. The study sponsors had no role in the design of the study; the collection, analysis, and interpretation of the data; the writing of the manuscript; or the decision to submit the manuscript for publication.
The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Neurological Disorders and Stroke or the National Institutes of Health. The complete list of WHI centers and investigators can be found online at http://www.whiscience.org/publications/WHI_investigators_shortlist.pdf.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Disclosures
None
REFERENCES
- 1.Zarnegar R, Michalopoulos G. Purification and biological characterization of human hepatopoietin a, a polypeptide growth factor for hepatocytes. Cancer Res. 1989;49:3314–3320. [PubMed] [Google Scholar]
- 2.Nakamura T, Nawa K, Ichihara A. Partial purification and characterization of hepatocyte growth factor from serum of hepatectomized rats. Biochem Biophys Res Commun. 1984;122:1450–1459. doi: 10.1016/0006-291x(84)91253-1. [DOI] [PubMed] [Google Scholar]
- 3.Boros P, Miller CM. Hepatocyte growth factor: A multifunctional cytokine. Lancet. 1995;345:293–295. doi: 10.1016/s0140-6736(95)90279-1. [DOI] [PubMed] [Google Scholar]
- 4.Morishita R, Aoki M, Yo Y, Ogihara T. Hepatocyte growth factor as cardiovascular hormone: Role of hgf in the pathogenesis of cardiovascular disease. Endocr J. 2002;49:273–284. doi: 10.1507/endocrj.49.273. [DOI] [PubMed] [Google Scholar]
- 5.Yamamoto Y, Kohara K, Tabara Y, Igase M, Nakura J, Miki T. Plasma hepatocyte growth factor and the relationship between risk factors and carotid atherosclerosis. Hypertens Res. 2002;25:661–667. doi: 10.1291/hypres.25.661. [DOI] [PubMed] [Google Scholar]
- 6.Satani K, Konya H, Hamaguchi T, Umehara A, Katsuno T, Ishikawa T, Kohri K, Hasegawa Y, Suehiro A, Kakishita E, Namba M. Clinical significance of circulating hepatocyte growth factor, a new risk marker of carotid atherosclerosis in patients with type 2 diabetes. Diabet Med. 2006;23:617–622. doi: 10.1111/j.1464-5491.2006.01849.x. [DOI] [PubMed] [Google Scholar]
- 7.Kawamoto R, Oka Y, Yoshida O, Takagi Y. Significance of serum circulating hepatocyte growth factor in the development of carotid atherosclerosis. J Atheroscler Thromb. 2003;10:154–159. doi: 10.5551/jat.10.154. [DOI] [PubMed] [Google Scholar]
- 8.Liu Y, Wilkinson FL, Kirton JP, Jeziorska M, Iizasa H, Sai Y, Nakashima E, Heagerty AM, Canfield AE, Alexander MY. Hepatocyte growth factor and c-met expression in pericytes: Implications for atherosclerotic plaque development. J Pathol. 2007;212:12–19. doi: 10.1002/path.2155. [DOI] [PubMed] [Google Scholar]
- 9.Rehman J, Considine RV, Bovenkerk JE, Li J, Slavens CA, Jones RM, March KL. Obesity is associated with increased levels of circulating hepatocyte growth factor. J Am Coll Cardiol. 2003;41:1408–1413. doi: 10.1016/s0735-1097(03)00231-6. [DOI] [PubMed] [Google Scholar]
- 10.Nakamura S, Morishita R, Moriguchi A, Yo Y, Nakamura Y, Hayashi S, Matsumoto K, Nakamura T, Higaki J, Ogihara T. Hepatocyte growth factor as a potential index of complication in diabetes mellitus. J Hypertens. 1998;16:2019–2026. doi: 10.1097/00004872-199816121-00025. [DOI] [PubMed] [Google Scholar]
- 11.Hiratsuka A, Adachi H, Fujiura Y, Yamagishi S, Hirai Y, Enomoto M, Satoh A, Hino A, Furuki K, Imaizumi T. Strong association between serum hepatocyte growth factor and metabolic syndrome. J Clin Endocrinol Metab. 2005;90:2927–2931. doi: 10.1210/jc.2004-1588. [DOI] [PubMed] [Google Scholar]
- 12.Matsumori A, Takano H, Obata JE, Takeda S, Tsuyuguchi N, Ono K, Okada M, Miyamoto T, Ohnishi T, Daikuhara Y, Sasayama S. Circulating hepatocyte growth factor as a diagnostic marker of thrombus formation in patients with cerebral infarction. Circ J. 2002;66:216–218. doi: 10.1253/circj.66.216. [DOI] [PubMed] [Google Scholar]
- 13.Anan F, Shimomura T, Imagawa M, Masaki T, Nawata T, Takahashi N, Yonemochi H, Eshima N, Saikawa T, Yoshimatsu H. Predictors for silent cerebral infarction in patients with chronic renal failure undergoing hemodialysis. Metabolism. 2007;56:593–598. doi: 10.1016/j.metabol.2007.01.003. [DOI] [PubMed] [Google Scholar]
- 14.Hays J, Hunt JR, Hubbell FA, Anderson GL, Limacher M, Allen C, Rossouw JE. The women's health initiative recruitment methods and results. Ann Epidemiol. 2003;13:S18–S77. doi: 10.1016/s1047-2797(03)00042-5. [DOI] [PubMed] [Google Scholar]
- 15.Langer RD, White E, Lewis CE, Kotchen JM, Hendrix SL, Trevisan M. The women's health initiative observational study: Baseline characteristics of participants and reliability of baseline measures. Ann Epidemiol. 2003;13:S107–S121. doi: 10.1016/s1047-2797(03)00047-4. [DOI] [PubMed] [Google Scholar]
- 16.The Women's Health Initiative Study Group. Design of the women's health initiative clinical trial and observational study. Control Clin Trials. 1998;19:61–109. doi: 10.1016/s0197-2456(97)00078-0. [DOI] [PubMed] [Google Scholar]
- 17.Kaplan RC, McGinn AP, Baird AE, Hendrix SL, Kooperberg C, Lynch J, Rosenbaum DM, Johnson KC, Strickler HD, Wassertheil-Smoller S. Inflammation and hemostasis biomarkers for predicting stroke in postmenopausal women: The women's health initiative observational study. J Stroke Cerebrovasc Dis. 2008;17:344–355. doi: 10.1016/j.jstrokecerebrovasdis.2008.04.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Wassertheil-Smoller S, Kooperberg C, McGinn AP, Kaplan RC, Hsia J, Hendrix SL, Manson JE, Berger JS, Kuller LH, Allison MA, Baird AE. Lipoprotein-associated phospholipase a2, hormone use, the risk of ischemic stroke in postmenopausal women. Hypertension. 2008;51:1115–1122. doi: 10.1161/HYPERTENSIONAHA.107.103721. [DOI] [PubMed] [Google Scholar]
- 19.Adams HP, Jr, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, Marsh EE., 3rd Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. Toast. Trial of org 10172 in acute stroke treatment. Stroke. 1993;24:35–41. doi: 10.1161/01.str.24.1.35. [DOI] [PubMed] [Google Scholar]
- 20.Luminex. Xmap technology. 2009 [Google Scholar]
- 21.Kaplan RC, Ho GY, Xue X, Rajpathak S, Cushman M, Rohan TE, Strickler HD, Scherer PE, Anastos K. Within-individual stability of obesity-related biomarkers among women. Cancer Epidemiol Biomarkers Prev. 2007;16:1291–1293. doi: 10.1158/1055-9965.EPI-06-1089. [DOI] [PubMed] [Google Scholar]
- 22.Haug C, Schmid-Kotsas A, Zorn U, Bachem MG, Schuett S, Gruenert A, Rozdzinski E. Hepatocyte growth factor is upregulated by low-density lipoproteins and inhibits endothelin-1 release. Am J Physiol Heart Circ Physiol. 2000;279:H2865–H2871. doi: 10.1152/ajpheart.2000.279.6.H2865. [DOI] [PubMed] [Google Scholar]
- 23.Moulton KS, Vakili K, Zurakowski D, Soliman M, Butterfield C, Sylvin E, Lo KM, Gillies S, Javaherian K, Folkman J. Inhibition of plaque neovascularization reduces macrophage accumulation and progression of advanced atherosclerosis. Proc Natl Acad Sci U S A. 2003;100:4736–4741. doi: 10.1073/pnas.0730843100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Beilmann M, Vande Woude GF, Dienes HP, Schirmacher P. Hepatocyte growth factor-stimulated invasiveness of monocytes. Blood. 2000;95:3964–3969. [PubMed] [Google Scholar]
- 25.Ma H, Calderon TM, Fallon JT, Berman JW. Hepatocyte growth factor is a survival factor for endothelial cells and is expressed in human atherosclerotic plaques. Atherosclerosis. 2002;164:79–87. doi: 10.1016/s0021-9150(02)00062-x. [DOI] [PubMed] [Google Scholar]
- 26.Takiuchi S, Mannami T, Miyata T, Kamide K, Tanaka C, Kokubo Y, Koyama Y, Inamoto N, Katsuya T, Iwai N, Kawano Y, Ogihara T, Tomoike H. Identification of 21 single nucleotide polymorphisms in human hepatocyte growth factor gene and association with blood pressure and carotid atherosclerosis in the japanese population. Atherosclerosis. 2004;173:301–307. doi: 10.1016/j.atherosclerosis.2003.12.020. [DOI] [PubMed] [Google Scholar]
- 27.Bell LN, Ward JL, Degawa-Yamauchi M, Bovenkerk JE, Jones R, Cacucci BM, Gupta CE, Sheridan C, Sheridan K, Shankar SS, Steinberg HO, March KL, Considine RV. Adipose tissue production of hepatocyte growth factor contributes to elevated serum hgf in obesity. Am J Physiol Endocrinol Metab. 2006;291:E843–E848. doi: 10.1152/ajpendo.00174.2006. [DOI] [PubMed] [Google Scholar]
- 28.Fain JN, Madan AK, Hiler ML, Cheema P, Bahouth SW. Comparison of the release of adipokines by adipose tissue, adipose tissue matrix, and adipocytes from visceral and subcutaneous abdominal adipose tissues of obese humans. Endocrinology. 2004;145:2273–2282. doi: 10.1210/en.2003-1336. [DOI] [PubMed] [Google Scholar]
- 29.Rahimi N, Saulnier R, Nakamura T, Park M, Elliott B. Role of hepatocyte growth factor in breast cancer: A novel mitogenic factor secreted by adipocytes. DNA Cell Biol. 1994;13:1189–1197. doi: 10.1089/dna.1994.13.1189. [DOI] [PubMed] [Google Scholar]
- 30.Vistoropsky Y, Trofimov S, Malkin I, Kobyliansky E, Livshits G. Genetic and environmental determinants of hepatocyte growth factor levels and their association with obesity and blood pressure. Ann Hum Biol. 2008;35:93–103. doi: 10.1080/03014460701822003. [DOI] [PubMed] [Google Scholar]
- 31.Anan F, Masaki T, Yonemochi H, Takahashi N, Nakagawa M, Eshima N, Saikawa T, Yoshimatsu H. Hepatocyte growth factor levels are associated with the results of 123i-metaiodobenzylguanidine myocardial scintigraphy in patients with type 2 diabetes mellitus. Metabolism. 2009;58:167–173. doi: 10.1016/j.metabol.2008.09.009. [DOI] [PubMed] [Google Scholar]
- 32.Lieb W, Safa R, Benjamin EJ, Xanthakis V, Yin X, Sullivan LM, Larson MG, Smith HM, Vita JA, Mitchell GF, Sawyer DB, Vasan RS. Vascular endothelial growth factor, its soluble receptor, and hepatocyte growth factor: Clinical and genetic correlates and association with vascular function. Eur Heart J. 2009 doi: 10.1093/eurheartj/ehp007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Nakamura Y, Morishita R, Nakamura S, Aoki M, Moriguchi A, Matsumoto K, Nakamura T, Higaki J, Ogihara T. A vascular modulator, hepatocyte growth factor, is associated with systolic pressure. Hypertension. 1996;28:409–413. doi: 10.1161/01.hyp.28.3.409. [DOI] [PubMed] [Google Scholar]
- 34.Morishita R, Moriguchi A, Higaki J, Ogihara T. Hepatocyte growth factor (hgf) as a potential index of severity of hypertension. Hypertens Res. 1999;22:161–167. doi: 10.1291/hypres.22.161. [DOI] [PubMed] [Google Scholar]
- 35.Nakamura S, Moriguchi A, Morishita R, Aoki M, Yo Y, Hayashi S, Nakano N, Katsuya T, Nakata S, Takami S, Matsumoto K, Nakamura T, Higaki J, Ogihara T. A novel vascular modulator, hepatocyte growth factor (hgf), as a potential index of the severity of hypertension. Biochem Biophys Res Commun. 1998;242:238–243. doi: 10.1006/bbrc.1997.7800. [DOI] [PubMed] [Google Scholar]