Sage Journals: Discover world-class research

Abstract

The tridecapeptide neurotensin has been implicated in the pathogenesis of cardiometabolic disease. Its stable precursor, pro-neurotensin/neuromedin N (pro-NT/NMN), has been associated with composite cardiovascular outcomes including coronary heart disease (CHD) and stroke. The exclusive association of pro-NT/NMN with ischemic stroke has not been evaluated. We conducted a prospective case-cohort study in the REasons for Geographic And Racial Differences in Stroke (REGARDS) study. From 2003 to 2007, REGARDS enrolled 30,239 white or black adults aged ⩾ 45 years. Baseline fasting pro-NT/NMN was measured by immunoassay in the analytic sample including 448 incident ischemic stroke cases and 818 random cohort sample participants. A total of 464 ischemic strokes occurred. Risk of stroke was assessed with a Cox proportional-hazards model incorporating demographic covariates and a second adding stroke risk factors. Increased pro-NT/NMN was associated with ischemic stroke in the demographic model overall (hazard ratio (HR) per standard deviation (SD) pro-NT/NMN 1.16, 95% confidence interval (CI) 1.01–1.33) and in men (HR per SD pro-NT/NMN 1.25, 95% CI 1.04–1.50); HRs were attenuated in the risk factor model. Pre-existing diabetes mellitus and CHD were the largest confounders of ischemic stroke risk, each accounting for an estimated 19% of the association of pro-NT/NMN with ischemic stroke observed in the demographic model. There were no significant interactions of race or sex with pro-NT/NMN. Further research on associations of pro-NT/NMN with stroke risk factors such as diabetes mellitus is indicated.

Keywords

biomarkers cohort studies neuropeptides neurotensin stroke

Introduction

Stroke and cerebrovascular disease have an enormous global burden¹ and considerable variation by geography,² race,³ and sex.⁴ Conditions such as diabetes mellitus, coronary heart disease (CHD), and hypertension have long been recognized to augment stroke risk.⁵ Neurotensin, a tridecapeptide initially isolated from the central nervous system,⁶ is present in relatively massive proportions in the periphery,⁷ where it has important effects on metabolism, including insulin resistance and obesity,⁸ and may be involved in regulation of blood pressure.⁹ Measurement of its stable precursor, pro-neurotensin/neuromedin N (pro-NT/NMN), in humans has implicated the neurotensin-related peptides in risk of obesity,⁸ diabetes mellitus,¹⁰ and composite incident cardiovascular disease (CVD) outcomes including CHD events and stroke.^10–12 Some associations of pro-NT/NMN with diabetes mellitus^10,11 and CVD¹⁰ were specific to or had a greater magnitude in women, suggesting the particular importance of neurotensin in women’s cardiometabolic health.¹³

That pro-NT/NMN was not associated with incident CHD events in a recent study from the REasons for Geographic And Racial Differences in Stroke (REGARDS) cohort¹⁴ makes it likely that previously reported associations of pro-NT/NMN with incident composite CVD are driven by an association of pro-NT/NMN with stroke, which has not been exclusively studied. While the risks of CHD and stroke are correlated, there are differences in risk factors, such as the weak association of lipid fraction levels with stroke despite a strong association with CHD.¹⁵ Hence, individuals may have substantially different risk for these two conditions.¹⁶ The purpose of this investigation was to assess the association of pro-NT/NMN with ischemic stroke in REGARDS, a modern, biracial cohort.

Methods

Sample

The REGARDS study was previously described.¹⁷ Briefly, REGARDS enrolled 30,239 black or white participants aged ⩾ 45 years from 2003 to 2007. Black Americans and a region of the southeastern US known as the Stroke Belt were oversampled. Immediate exclusions included insufficient English proficiency, apparent cognitive impairment on telephone interview, active cancer treatment, residence in or waitlisting for a nursing home, and medical conditions precluding long-term follow-up. Medical history and verbal consent were obtained during the initial telephone interview, followed by an in-home visit at which written consent, medication inventory, biometrics, electrocardiogram (EKG), fasting phlebotomy, and urine samples were obtained. The methods of the REGARDS study have been approved by the institutional review board of each involved institution.

Participants or their representatives are surveyed every 6 months by telephone to assess for the presence of stroke occurrence or symptoms. Reports of possible stroke, TIA, or death were screened by a stroke nurse, validated by medical record review, and adjudicated by a committee of stroke experts. Validation of stroke cases required focal neurologic symptoms of ⩾ 24 hours’ duration or nonfocal neurologic symptoms in conjunction with imaging findings suggesting stroke.¹⁸ Stroke events were initially classified as ischemic or hemorrhagic, with ischemic strokes further subclassified according to the TOAST criteria.¹⁹

For this analysis, a random cohort sample of 1127 participants, stratified on age, race, and sex, was drawn from the full REGARDS cohort.²⁰ In addition to this subcohort, 553 adjudicated cases of incident ischemic stroke occurring prior to September 1, 2011 were identified and added to the random cohort sample to form a raw case-cohort²¹ analytic sample. Participants were excluded from analysis based on the criteria shown in Figure 1. Those excluded for being nonfasting were similar to those who were fasting, except they were more likely to have diabetes mellitus or atrial fibrillation (data not shown).

Figure 1.

Exclusions of participants from analysis.

Variables

Self-reported demographic variables included age (years), sex, race, education (less than high school graduate, high school graduate, some college, or ⩾ college degree), annual income (< $20,000, $20,000 – $34,999, $35,000 – $74,999, ⩾ $75,000, or refused), geographic region²² (Stroke Belt, Stroke Buckle, or other), tobacco smoking in pack-years or baseline status (never, current, or past smoker), alcohol use (none, moderate [1–7 alcoholic drinks per week in women, 1–14 alcoholic drinks per week in men], or heavy [> 7 alcoholic drinks per week in women, > 14 alcoholic drinks per week in men]), weekly exercise frequency (none, 1–3 times per week, ⩾ 4 times per week; each occurrence was defined as intense physical activity sufficient to work up a sweat).

Clinical variables included history of stroke (self-reported physician diagnosis), diabetes mellitus (self-reported use of antihyperglycemic medications or insulin or fasting glucose ⩾ 126 mg/dL), CHD (self-reported physician diagnosis of myocardial infarction, coronary revascularization procedure, coronary artery bypass graft surgery, or EKG findings to suggest prior infarct), atrial fibrillation (AF; self-reported physician diagnosis or EKG finding), left ventricular hypertrophy (LVH; EKG finding using Sokolow-Lyon criteria²³), hypertension (self-reported use of antihypertensive medications, systolic blood pressure (SBP) ⩾ 130 mmHg, or diastolic blood pressure (DBP) ⩾ 80 mmHg).²⁴ Body mass index (BMI), in kg/m², was calculated from baseline height and weight measurements.

Laboratory variables

Fasting blood samples were obtained at the in-home visit, centrifuged locally for 10 minutes, then shipped on ice to the central laboratory at the University of Vermont, where they were re-centrifuged and stored at –80°C. Glucose was measured using colorimetric reflectance spectrophotometry with the Ortho Vitros 950 IRC Clinical Analyzer (Johnson & Johnson Clinical Diagnostics, New Brunswick, NJ, USA). Pro-NT/NMN was measured in the random cohort and stroke cases from thawed EDTA plasma using a single-step sandwich immunoassay (SphingoTec GmbH, Hennigsdorf, Germany) based on a coated tube technique and chemiluminescence label. The assay has a functional sensitivity of 10 pmol/L, detection limit of 1.9 pmol/L, and interassay precision of 20% of the coefficient of variation (5–9%). Pro-NT/NMN assays were performed at an independent laboratory (ICI Immunochemical Intelligence GmbH, Berlin, Germany) by technicians who were blinded to clinical data.

Statistical analysis

All analyses were weighted to account for case-cohort design, using a Taylor series as a finite population correction, and were two-sided with α = 0.05 and interaction α = 0.10. Pro-NT/NMN was positively skewed and was thus common logarithm-transformed for continuous analyses or quartiled for other analyses. Baseline variables were tabulated by pro-NT/NMN quartiles, with Pearson chi-squared tests with second-order correction according to Rao and Scott²⁵ used to compare categorical variables and simple linear regression used to compare continuous variables.

Two multivariable Cox proportional-hazards models²⁶ were fitted to examine the association of baseline pro-NT/NMN with ischemic stroke. Model 1 included demographic variables: age, sex, region, income, education, and an age*race multiplicative interaction term (previously shown to be significant in REGARDS).²⁷ Model 2 added Framingham Stroke risk factors: SBP, antihypertensive drug use, smoking, baseline diabetes mellitus, CHD, AF, and LVH.²⁸ The association of log pro-NT/NMN with ischemic stroke was presented visually with sex-stratified restricted cubic splines, with prespecified knots determined by Harrell’s method²⁹ and referencing the sex-specific median. Kernel density plots show distributions of log pro-NT/NMNN stratified by sex.

Stroke-free survival time was measured from date of phlebotomy (in-home visit) to date of stroke or censoring. Those who did not develop stroke in follow-up were censored on the earliest date of: last follow-up, death, or September 1, 2011 (index date of case-cohort creation). Failure date was the adjudicated date of stroke cases. The interaction of sex and race with log pro-NT/NMN was tested with multiplicative interaction terms.

An attenuation analysis was planned to examine the potential confounding of stroke risk factors (Model 2) on an overall or group-specific association of log pro-NT/NMN with stroke. Model 1 covariates were used as a base, yielding HR_{Model 1}. Each Model 2 covariate was individually and separately added to this base model, producing HR_{Model 1 + Risk Factor}. The estimated attenuation of a Model 1 association of pro-NT/NMN with stroke was calculated accordingly:

$%_{atten} {= (HR}_{Model 1} - {HR}_{Model 1} +_{Risk Factor}) /$ ${(HR}_{Model 1} - 1) \times 100$

Statistical analysis was conducted with Stata, version 16.0 (StataCorp, College Station, TX, USA).

Results

Sample characteristics

A total of 1266 participants were included following exclusion criteria shown in Figure 1: 818 from the cohort random sample (including 16 incident ischemic strokes), and 448 incident ischemic stroke cases from the remainder of the REGARDS cohort. Thus, there were 464 ischemic strokes occurring over the median follow-up time of 4.7 years (IQR 2.5–6.4 years). Of the incident ischemic strokes, 94 (20.3%) were cardioembolic, 78 (16.9%) were small-vessel, 59 (12.7%) were large-vessel, and 233 (50.2%) were of a miscellaneous but known cause or had an unclassifiable cause based on available data.

Baseline characteristics

In the cohort random sample, pro-NT/NMN ranged from 21 pmol/L to 1278 pmol/L, with median 169 pmol/L and interquartile range 105–273 pmol/L. Table 1 shows baseline characteristics of cohort random sample participants across pro-NT/NMN quartiles. Higher quartiles of pro-NT/NMN had larger proportions of participants who were less educated, of a lower annual income category, not current alcohol users, and had pre-existing diabetes mellitus or hypertension (p < 0.05).

Table 1.

Baseline characteristics of cohort random sample^a by pro-NT/NMN quartile (n = 818).

Pro-NT/NMN quartiles	Q1(21–105 pmol/L)	Q2(105–169 pmol/L)	Q3(169–273 pmol/L)	Q4(273–5441 pmol/L)	p
Age, years; mean (95% CI)	71.9 (68.9, 74.9)	73.9 (71.0, 76.8)	70.1 (66.8, 73.4)	73.0 (70.1, 75.9)	0.99^b
Black race (%)	46.3	56.4	67.9	70.5	< 0.001^c
Male sex (%)	62.4	58.2	54.3	41.8	0.72^c
Region (%)
Stroke Belt	28.5	33.3	31.6	35.1	0.83^c
Stroke Buckle	22.3	15.2	19.6	15.3
Other	49.2	51.5	48.8	49.6
Annual income (%)
< $20,000	8.3	30.4	21.8	25.8	0.01^c
$20,000 – $34,999	25.2	27.7	32.2	22.3
$35,000 – $74,999	21.2	20.8	19.6	24.4
⩾ $75,000	19.0	9.2	13.1	8.1
Refused	16.4	11.9	13.4	19.4
Education (%)
< High school	15.8	26.4	12.1	26.9	0.03^c
High school graduate	15.9	20.5	28.5	20.2
Some college	26.1	21.9	16.5	26.8
⩾ College graduate	42.3	31.2	42.9	26.2
Weekly exercise frequency (%)
None	30.6	41.0	40.8	40.5	0.03^c
1–3 times	37.2	42.9	29.0	26.2
⩾ 4 times	32.2	16.2	30.2	33.4
Tobacco (%)
Never smoker	51.6	47.8	55.7	42.0	0.27^c
Past smoker	41.0	39.3	32.3	41.9
Current smoker	7.4	12.8	11.9	16.1
Current alcohol use (%)
None	60.3	63.3	67.7	80.7	0.01^c
Moderate	32.3	34.9	28.5	17.6
Heavy	7.4	1.9	3.8	1.7
Diabetes mellitus (%)	0.7	1.2	2.8	6.8	0.02^c
Coronary heart disease (%)	18.3	18.4	21.0	21.6	0.91^c
Hypertension (%)	68.0	81.3	78.8	87.9	< 0.001^c
BMI category (%)
< 18.5 Underweight	0.9	0.0	2.1	0.6	0.55^c
18.5–24.9 Normal weight	35.1	37.1	28.6	30.2
25.0–29.9 Overweight	38.0	33.9	39.5	33.0
⩾ 30.0 Obesity	26.0	29.0	29.8	37.2
Left ventricular hypertrophy (%)	8.7	17.4	6.5	11.7	0.13^c
Atrial fibrillation (%)	5.9	10.0	11.4	4.4	0.12^c

Weighted to analytic cohort; ^blinear regression; ^cRao–Scott chi-squared test.

BMI, body mass index; pro-NT/NMN, pro-neurotensin/neuromedin N.

Association of pro-NT/NMN with incident ischemic stroke

Table 2 shows an association of increased pro-NT/NMN with risk of ischemic stroke in Model 1 (HR per SD log pro-NT/NMN 1.16, 95% CI 1.01–1.33; 4^th versus 1^st quartile, HR 1.49, 95% CI 1.04–2.15), with a significant trend of increased risk across successive quartiles (p = 0.03). This association was not maintained with further adjustment for stroke risk factors.

Table 2.

Hazard ratio (95% CI) for ischemic stroke by pro-NT/NMN, overall and stratified by sex.^a

	n	Per SD log pro-NT/NMN	Quartiles
		Per SD log pro-NT/NMN		Q1	Q2	Q3	Q4	p for trend
Overall		Model 1	464	1.16 (1.01, 1.33)	1 (ref)	1.04 (0.75, 1.44)	1.19 (0.87, 1.63)	1.49 (1.04, 2.15)	0.03
Overall	Model 2	1.11 (0.97, 1.27)	464	1 (ref)	0.78 (0.54, 1.13)	0.97 (0.69, 1.36)	1.20 (0.84, 1.72)	0.16
Women	Model 1	220	1.08 (0.87, 1.35)	1 (ref)	1.30 (0.76, 2.24)	1.32 (0.82, 2.14)	1.37 (0.77, 2.42)	0.33
Women	Model 2	220	1.14 (0.93, 1.40)	1 (ref)	1.29 (0.71, 2.34)	1.32 (0.78, 2.24)	1.44 (0.79, 2.65)	0.24
Men	Model 1	244	1.25 (1.04, 1.50)	1 (ref)	0.92 (0.60, 1.42)	1.17 (0.77, 1.80)	1.71 (1.06, 2.76)	0.02
Men	Model 2	244	1.17 (0.98, 1.40)	1 (ref)	0.66 (0.41, 1.09)	0.88 (0.55, 1.41)	1.31 (0.83, 2.06)	0.14

Covariates in multivariable models (sex and race removed as covariates when used in stratification): Model 1: age, sex, race, region, income, education, age*race interaction term; Model 2: Model 1 covariates and systolic blood pressure, antihypertensive therapy, prevalent diabetes mellitus, smoking (pack-years), prevalent coronary heart disease, atrial fibrillation, left ventricular hypertrophy.

pro-NT/NMN*sex Model 1 interaction p = 0.23.

CI, confidence interval; HR, hazard ratio; pro-NT/NMN, pro-neurotensin/neuromedin N; SD, standard deviation.

Similarly, increased pro-NT/NMN was associated with ischemic stroke in men with minimal adjustment (HR per SD log pro-NT/NMN 1.25, 95% CI 1.04–1.50; 4^th versus 1^st quartile, HR 1.71, 95% CI 1.06–2.76), with a significant trend across quartiles (p = 0.02); this association was attenuated in the full model. No association of pro-NT/NMN with ischemic stroke was observed in women; the apparent sex difference was not statistically significant (interaction p = 0.23 in Model 1). The online supplemental Figure 1 shows sex-stratified restricted cubic splines of the association of pro-NT/NMN with incident ischemic stroke. The association of pro-NT/NMN with ischemic stroke did not differ by race (interaction p = 0.41 in Model 2).

Online supplemental Table 1 reports the associations of pro-NT/NMN with ischemic stroke subtypes. Pro-NT/NMN was not associated with a specific risk of small-vessel, large-vessel, or cardioembolic stroke, but an association of log pro-NT/NMN with miscellaneous or unclassified stroke was observed in the fully adjusted model (HR per SD log pro-NT/NMN 1.22, 95% CI 1.03–1.44).

Attenuation analysis of association of pro-NT/NMN with ischemic stroke

As shown in Table 3, pre-existing diabetes mellitus and CHD were the most prominent confounders of the association of log pro-NT/NMN with incident stroke, each accounting for an estimated 19% of the per SD log pro-NT/NMN risk of ischemic stroke. For men, CHD at baseline was the most prominent confounder, accounting for an estimated 12% of the per SD log pro-NT/NMN hazard of ischemic stroke.

Table 3.

Covariate attenuation of the association of log pro-NT/NMN with ischemic stroke in Model 1.

	Overall			Men
	HR per SD log pro-NT/NMN	95% CI	% Attenuation	HR per SD log pro-NT/NMN	95% CI	% Attenuation
Model 1 alone	1.16	(1.01, 1.33)		1.25	(1.04, 1.50)
SBP	1.14	(0.99, 1.31)	13%	1.24	(1.04, 1.47)	4%
Antihypertensive use	1.14	(1.00, 1.31)	13%	1.23	(1.02, 1.45)	8%
Diabetes mellitus	1.13	(0.97, 1.31)	19%	1.23	(1.02, 1.47)	8%
Smoking	1.17	(1.03, 1.34)	−6%	1.23	(1.03, 1.47)	8%
AF	1.16	(1.01, 1.33)	0%	1.25	(1.06, 1.49)	0%
LVH	1.15	(1.01, 1.32)	6%	1.23	(1.03, 1.47)	8%
CHD	1.13	(0.99, 1.31)	19%	1.22	(1.02, 1.47)	12%

AF, atrial fibrillation; CHD, coronary heart disease; CI, confidence interval; HR, hazard ratio; LVH, left ventricular hypertrophy; pro-NT/NMN, pro-neurotensin/neuromedin N; SBP, systolic blood pressure; SD, standard deviation.

Discussion

In this prospective case-cohort study within the modern and biracial REGARDS study, higher baseline plasma pro-NT/NMN was associated with increased risk of ischemic stroke overall and in men when controlling for participant demographics. Specifically, participants with a 4^th versus 1^st-quartile pro-NT/NMN had a 49% increased ischemic stroke risk overall; men had a 71% increased risk when controlling for demographics. These associations were attenuated after controlling for traditional stroke risk factors. Together, these findings suggest that pro-NT/NMN augments stroke risk – particularly in men – through its influence on traditional stroke risk factors such as diabetes mellitus and CHD.

Our findings are consistent with reports of three previous epidemiologic studies, which found a 10–24% increased risk of composite CVD per SD higher pro-NT/NMN.^10–12 However, unlike previously documented women-specific or larger-magnitude associations of pro-NT/NMN in women with composite CVD^10,12 and diabetes mellitus,^10,11 we did not observe similar sex differences in the association of pro-NT/NMN with ischemic stroke. Rather, we observed a 25% increased risk of ischemic stroke per SD log pro-NT/NMN in men but no association in women, although the difference in associations by sex was not statistically significant. We find it most probable that this discordance is due to the exclusive study of stroke in the present investigation. This notion underscores that, while stroke and myocardial infarction share some common risk factors, they are separate disease processes with heterogenous etiologies. Thus, their separate consideration as outcomes in human epidemiologic studies may improve validity.

We expected to find some differences in associations of pro-NT/NMN with subtypes of ischemic stroke. Surprisingly, and unlike N-terminal pro-B-type natriuretic peptide,³⁰ pro-NT/NMN was not associated with cardioembolic stroke. Together with the observed lack of association of pro-NT/NMN with atrial fibrillation, findings suggest pro-NT/NMN does not reflect atriopathy leading to stroke.

The association of pro-NT/NMN with ischemic stroke is plausibly related to the influence of the neurotensinergic peptides and their receptors on inflammation³¹ and metabolism. Pro-NT/NMN was previously associated with incident diabetes mellitus,^10,11 which substantially confounded the association of pro-NT/NMN with stroke in this study. This may be rooted in insulin resistance occurring as a result of neurotensin-driven augmentation of dietary fatty acid absorption.^8,32 Variation at the SORT1 gene, encoding a receptor to which neurotensin is a ligand, has been critically linked to CVD risk³³ and insulin resistance.³⁴ Accordingly, broader exploration of the interplay between pro-NT/NMN, diabetes mellitus, and dietary patterns in human subjects is necessary to fully explain these findings. As SBP and use of antihypertensive drugs also appreciably confounded the association of pro-NT/NMN with stroke, investigation of the association of pro-NT/NMN with hypertension is similarly warranted.

Study limitations and strengths

Several limitations of this study must be considered. Firstly, the generalizability of findings to race groups other than black or white is limited. Participants excluded for not fasting prior to phlebotomy were more likely to have atrial fibrillation and diabetes mellitus at baseline; omission bias might have occurred. Such a bias would likely push findings toward the null, except in the setting that excluded participants, like those with more severe diabetes mellitus who did not fast, might have different relationships of pro-NT/NMN with stroke. That a large number of ischemic strokes were not classifiable into subtype is a potential limitation of the subtype-stratified results. These strokes likely represent the range of stroke etiologies and were not classified primarily due to lack of a definite etiology as documented in medical records. Baseline characteristics and pro-NT/NMN levels did not differ among miscellaneous or unclassified stroke cases and the other ischemic stroke cases (data not shown). Lastly, pro-NT/NMN was measured at varied times from participants’ individual phlebotomy dates, raising concern for time-related breakdown of the molecule in storage, but this is unlikely given the general stability of peptides in frozen storage at –80°C or colder.³⁵

This study has several strengths. REGARDS is a large, prospective, and population-based cohort with ongoing follow-up and high retention (87.3% cumulative retention).³⁰ While previous cohort studies evaluating the association of pro-NT/NMN with aggregate CVD were ethnically homogenous,^36–38 REGARDS is the first study we are aware of to measure pro-NT/NMN in a significant proportion of non-white participants, improving on the generalizability of these findings. The rigorous screening and adjudicating of stroke cases by stroke experts minimized misclassification bias, although subtype could often not be determined. Semiannual monitoring for stroke and the addition of biometry and laboratory testing to participant self-report in establishing diagnoses of medical conditions further bolster the integrity of these data.

Conclusion

In this large and modern prospective cohort, increased baseline pro-NT/NMN was associated with increased risk of ischemic stroke when controlling for demographic factors. This finding may be specific to men, standing in contrast to prior investigations associating pro-NT/NMN with aggregate cardio- and cerebrovascular outcomes with findings specific to women. Pro-NT/NMN was not significantly associated with ischemic stroke when controlling for traditional stroke risk factors, and the less adjusted association was especially explained by the confounding effects of pre-existing diabetes mellitus and CHD. Further study of the relationship of pro-NT/NMN with incidence of stroke risk factors such as diabetes mellitus and hypertension is warranted.

Supplemental Material

ProNT_Supp_VascMed_18AUG – Supplemental material for Pro-neurotensin/neuromedin N and risk of ischemic stroke: The REasons for Geographic And Racial Differences in Stroke (REGARDS) study

Supplemental material, ProNT_Supp_VascMed_18AUG for Pro-neurotensin/neuromedin N and risk of ischemic stroke: The REasons for Geographic And Racial Differences in Stroke (REGARDS) study by Charles D Nicoli, Nicholas Wettersten, Suzanne E Judd, George Howard, Virginia J Howard, Joachim Struck and Mary Cushman in Vascular Medicine

Footnotes

Acknowledgements

The authors thank the other investigators, the staff, and the participants of the REGARDS study for their valuable contributions. A full list of participating REGARDS investigators and institutions can be found at: .

Declaration of conflicting interests

Dr Struck is employed by SphingoTec Gmbh, which funded measurement of pro-neurotensin/neuromedin N. The other authors report no disclosures.

Funding

This research is supported by cooperative agreement U01 NS041588 co-funded by the National Institute of Neurological Disorders and Stroke (NINDS) and the National Institute on Aging (NIA), National Institutes of Health, Department of Health and Human Services. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NINDS or the NIA. Representatives of the NINDS were involved in the review of the manuscript but were not directly involved in the collection, management, analysis or interpretation of the data.

ORCID iD

Charles D Nicoli

Supplemental material

The supplemental material is available online with the article.

References

Feigin

Norrving

Mensah

GA.

Global burden of stroke. Circ Res 2017; 120: 439–448.

Kim

Cahill

Cheng

NT.

Global stroke belt: Geographic variation in stroke burden worldwide. Stroke 2015; 46: 3564–3570.

Howard

VJ.

Ethnic disparities in stroke: The scope of the problem. Ethn Dis 2001; 11: 761–768.

Seshadri

Beiser

Kelly-Hayes

, et al. The lifetime risk of stroke: Estimates from the Framingham Study. Stroke 2006; 37: 345–350.

Wolf

D’Agostino

Belanger

, et al. Probability of stroke: A risk profile from the Framingham Study. Stroke 1991; 22: 312–318.

Carraway

Leeman

SE.

The isolation of a new hypotensive peptide, neurotensin, from bovine hypothalami. J Biol Chem 1973; 248: 6854–6861.

Carraway

Leeman

SE.

Characterization of radioimmunoassayable neurotensin in the rat. Its differential distribution in the central nervous system, small intestine, and stomach. J Biol Chem 1976; 251: 7045–7052.

Song

Zaytseva

, et al. An obligatory role for neurotensin in high-fat-diet-induced obesity. Nature 2016; 533: 411–415.

Rioux

Kérouac

St-Pierre

Analysis of the biphasic depressor-pressor effect and tachycardia caused by neurotensin in ganglion-blocked rats. Neuropeptides 1982; 3: 113–127.

10.

Melander

Maisel

Almgren

, et al. Plasma proneurotensin and incidence of diabetes, cardiovascular disease, breast cancer, and mortality. JAMA 2012; 308: 1469–1475.

11.

Fawad

Bergmann

Struck

, et al. Proneurotensin predicts cardiovascular disease in an elderly population. J Clin Endocrinol Metab 2018; 103: 1940–1947.

12.

Januzzi

Jr Lyass

Liu

, et al. Circulating proneurotensin concentrations and cardiovascular disease events in the community: The Framingham Heart Study. Arterioscler Thromb Vasc Biol 2016; 36: 1692–1697.

13.

Daniels

Maisel

AS.

Cardiovascular biomarkers and sex: The case for women. Nat Rev Cardiol 2015; 12: 588–596.

14.

Wettersten

Cushman

Howard

, et al. Usefulness of proneurotensin to predict cardiovascular and all-cause mortality in a United States population (from the Reasons for Geographic and Racial Differences in Stroke Study). Am J Cardiol 2018; 122: 26–32.

15.

Di Angelantonio

Sarwar

Perry

, et al. Major lipids, apolipoproteins, and risk of vascular disease. JAMA 2009; 302: 1993–2000.

16.

Howard

Cushman

Prineas

, et al. Advancing the hypothesis that geographic variations in risk factors contribute relatively little to observed geographic variations in heart disease and stroke mortality. Prev Med 2009; 49: 129–132.

17.

Howard

Cushman

Pulley

, et al. The Reasons for Geographic and Racial Differences in Stroke study: Objectives and design. Neuroepidemiology 2005; 25: 135–143.

18.

Howard

Kleindorfer

Judd

, et al. Disparities in stroke incidence contributing to disparities in stroke mortality. Ann Neurol 2011; 69: 619–627.

19.

Adams

Jr Bendixen

Kappelle

, et al. 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.

20.

Alexander

Zakai

Gillett

, et al. ABO blood type, factor VIII, and incident cognitive impairment in the REGARDS cohort. Neurology 2014; 83: 1271–1276.

21.

Prentice

RL.

A case-cohort design for epidemiologic cohort studies and disease prevention trials. Biometrika 1986; 73: 1–11.

22.

Howard

Anderson

Johnson

, et al. Evaluation of social status as a contributing factor to the stroke belt region of the United States. Stroke 1997; 28: 936–940.

23.

Sokolow

Lyon

. The ventricular complex in left ventricular hypertrophy as obtained by unipolar precordial and limb leads. Am Heart J 1949; 37(2): 161–186.

24.

Whelton

Carey

Aronow

, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults: A report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Hypertension 2018; 71: e13–e115.

25.

Rao

Scott

AJ.

The analysis of categorical data from complex sample surveys: Chi-squared tests for goodness of fit and independence in two-way tables. J Am Stat Assoc 1981; 76: 221–230.

26.

Cox

DR.

Regression models and life-tables. J R Stat Soc Series B Stat Methodol 1972; 34: 187–220.

27.

Howard

Cushman

Kissela

, et al. Traditional risk factors as the underlying cause of racial disparities in stroke: Lessons from the half-full (empty?) glass. Stroke 2011; 42: 3369–3375.

28.

D’Agostino

Wolf

Belanger

, et al. Stroke risk profile: Adjustment for antihypertensive medication. The Framingham Study. Stroke 1994; 25: 40–43.

29.

Harrell

. Ordinal logistic regression. In: Regression modeling strategies: with applications to linear models, logistic and ordinal regression, and survival analysis. Cham: Springer International Publishing, 2015, pp.311–325.

30.

Cushman

Judd

Howard

, et al. N-terminal pro-B-type natriuretic peptide and stroke risk: The reasons for geographic and racial differences in stroke cohort. Stroke 2014; 45: 1646–1650.

31.

Barchetta

Cimini

Capoccia

, et al. Neurotensin is a lipid-induced gastrointestinal peptide associated with visceral adipose tissue inflammation in obesity. Nutrients 2018; 10: 526.

32.

Armstrong

Parker

Ferris

, et al. Neurotensin stimulates [3H]oleic acid translocation across rat small intestine. Am J Physiol 1986; 251(6 Pt 1): G823–829.

33.

Musunuru

Strong

Frank-Kamenetsky

, et al. From noncoding variant to phenotype via SORT1 at the 1p13 cholesterol locus. Nature 2010; 466: 714–719.

34.

Rabinowich

Fishman

Hubel

, et al. Sortilin deficiency improves the metabolic phenotype and reduces hepatic steatosis of mice subjected to diet-induced obesity. J Hepatol 2015; 62: 175–181.

35.

Lewis

Callas

Jenny

, et al. Longitudinal stability of coagulation, fibrinolysis, and inflammation factors in stored plasma samples. Thromb Haemost 2001; 86: 1495–1500.

36.

Lahmann

Lissner

Gullberg

, et al. Differences in body fat and central adiposity between Swedes and European immigrants: The Malmö Diet and Cancer Study. Obes Res 2000; 8: 620–631.

37.

Berglund

Nilsson

Eriksson

, et al. Long-term outcome of the Malmö Preventive Project: Mortality and cardiovascular morbidity. J Intern Med 2000; 247: 19–29.

38.

Benjamin

Brown

Burke

, et al. American Heart Association Cardiovascular Genome-Phenome study: Foun-dational basis and program. Circulation 2015; 131: 100–112.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.15 MB