The association of neck circumference with incident congestive heart failure and coronary heart disease mortality in a community-based population with or without sleep-disordered breathing

The association of neck circumference with incident congestive heart failure and coronary heart... Background: Neck circumference (NC), representing upper body subcutaneous adipose tissue, may be correlated with increased risk of overweight/obesity, obstructive sleep apnoea, and metabolic and cardiovascular disease. However, the relationship between NC and the incidence of congestive heart failure (CHF) or mortality due to coronary heart disease (CHD) in a community-based population with and without sleep-disordered breathing (SDB) has not yet been clarified. Methods: We performed a prospective study using the Sleep Heart Health Study (SHHS) cohort. Cox proportional hazards regression models were used to estimate the association of different levels of NC with CHF incidence or CHD mortality in 2234 individuals with SDB and 2199 without SDB, respectively. Results: After adjusting for age, sex, and body mass index (BMI), NC was significantly associated with CHF when comparing the highest NC quartile group with the lowest (hazard ratio, HR, 2.265, 95% confidence interval, CI, 1.074–4.777) in the non-SDB population. This association diminished after further adjustment for other risk factors, but remained statistically significant, with an adjusted HR of 1.082 (95% CI 1.003–1.166) per unit increase in NC. Additionally, after adjustment for age, sex, and BMI, NC was also shown to be remarkably associated with CHD mortality (HR 1.141, 95% CI 1.014–1.282) per unit increase in NC in the non-SDB population but not in the SDB population. After adjustment for all the covariates, there was a significant association between NC and CHD death in those without SDB, with an adjusted HR of 1.134 (95% CI 1.001–1.284) per unit increase in NC. Conclusions: NC may correlate with CHF incidence and CHD mortality in population without SDB. NC measurement may help risk stratification for cardiovascular diseases. Trial registration: NCT00005275, January 1994. Keywords: Neck circumference, Congestive heart failure, Coronary heart disease, Mortality, Sleep-disordered breathing * Correspondence: gang_wang@mail.xjtu.edu.cn Jingjing Zhang and Qi Guo contributed equally to this work. Department of Emergency Medicine, the Second Affiliated Hospital of Xi’an Jiaotong University, Xi’an 710004, China Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Zhang et al. BMC Cardiovascular Disorders (2018) 18:108 Page 2 of 8 Background study, 1230 from the Cardiovascular Health Study, 688 Obesity, a pathological condition with body mass index from Framingham Offspring Cohort, and 1971 from (BMI) over 30 kg/m , is associated with a series of meta- other studies. Of them, 600 subjects were excluded be- bolic risk factors [1]. However, the overall obesity in- cause of pre-existing CHD, heart failure, or other related cludes unique intrinsic characteristics of different fat cardiovascular events, 762 because of lack of follow-up depots and the pattern of body fat distribution may also data, and 9 because of lack of NC data, ultimately leav- contribute to the prognosis [2]. Recently, significant het- ing an analytical sample size of 4433. SDB has been pro- erogeneity was found between visceral adipose tissue posed as a risk factor for cardiovascular events; besides, (VAT) and subcutaneous adipose tissue (SAT) of the SHHS mainly focused on the relationship between SDB upper body. VAT is recognized as a pathogenic fat de- and cardiovascular outcomes. Therefore, our investiga- posit with increased risk for insulin resistance, type II tion was carried out in the following two groups: the diabetes mellitus, and atherosclerosis [3, 4]. It was found group with SDB and the group without SDB (Additional that while lower body SAT may reduce the risk for car- file 1: Figure S1). diovascular disease, abdominal SAT was not related [5]. Previous studies have demonstrated that upper body NC and SDB fat produced the majority of systemic free fatty acids NC was measured during health interviews at the begin- (FFA), which was associated with insulin resistance, in- ning of the observation period. Participants were advised creased very low-density lipoprotein, and endothelial to sit upright and look straight ahead. An inelastic tape dysfunction [3]. Neck circumference (NC), measured at was applied around the neck just below the laryngeal the level of the laryngeal prominence, represents upper prominence. The NC measurement was made perpen- body SAT and may be correlated with increased risk of dicular to the long axis of the neck with the tape contact- overweight/obesity, obstructive sleep apnoea, type II dia- ing the skin surface under acceptable pressure. Subjects betes, metabolic syndrome, and cardiovascular disease were divided into NC quartiles of the following ranges: [3, 4, 6–10]. However, the relationship between NC and ≤34.1 (reference), 34.1–37.0, 37.0–40.5, and > 40.5 cm the incidence of congestive heart failure (CHF) or mor- [13]. Home polysomnography was performed with the tality due to coronary heart disease (CHD) has not yet Compumedics P-series portable monitor (Abbotsford, been clarified. Previously, the Sleep Heart Health Study Victoria, Australia). Detailed information were collected (SHHS) was carried out in a geographically diverse, as described in a previously published work [14]. Apnoea community-based population to assess the difference in was defined as complete or almost complete cessation of cardiovascular outcomes among participants with differ- airflow lasting for at least 10 s, while hypopnea was de- ent sleep apnoea statuses [11]. Based on SHHS, the fined as a clearly discernible decrease in airflow, or chest present study aimed to investigate the role of NC in the or abdominal plethysmograph amplitude that lasted for at incidence of CHF or CHD mortality in populations with least 10 s. Both, apnoea and hypopnea, required an as- and without sleep-disordered breathing (SDB). sociated 4% or greater oxyhaemoglobin desaturation. AHI is the average number of episodes of apnoea and/ Methods or hypopnea per hour of sleep [15]. SDB was defined as Study design AHI ≥ 5 respiratory events per hour while non-SDB was Analyses were carried out in two subsets of data ob- defined as AHI < 5. tained from the participants of SHHS (ClinicalTrials.gov Identifier: NCT00005275). A total of 5804 participants aged 40 years and older underwent home polysomnogra- CHD death and CHF phy and completed a set of questionnaires on general Baseline CHD or CHF was defined to be present if the health and sleep habits since 1994 [12]. Cardiovascular participant responded positively to a standardized ques- disease outcomes were tracked until 2010. Detailed aims tionnaire before the polysomnogram monitoring, or re- and design of SHHS are described elsewhere [11]. The ported that they had undergone coronary bypass surgery protocol was approved by the Institutional Review Board or coronary angioplasty, or if the parent cohort had an of each participating institution and signed informed identified CHD event or CHF before the SHHS baseline. consents were provided by the subjects. The data were CHD death was defined as fatal CHD at any time be- accessed based on a signed agreement with the Brigham tween the baseline polysomnogram and the final and Women’s Hospital. follow-up between 2008 and 2011. Incident CHF was de- fined as the first occurrence of heart failure during this Participants period of follow-up. Ongoing surveillance for CHD A total of 5804 subjects from SHHS cohort included deaths and incident CHF events was carried out by the 1915 from the Atherosclerosis Risk in Communities parent cohorts according to cohort-specific protocols. Zhang et al. BMC Cardiovascular Disorders (2018) 18:108 Page 3 of 8 Other covariates with SDB (P < 0.05) (Additional file 2:Table S1).More Smoking status was classified as “never” (if the partici- baseline characteristics are shown in Additional file 3: pant reported lifetime smoking of fewer than 20 packs Table S2. of cigarettes), “former”,or “current”. Educational level During the follow-up period, 163 CHF events and 58 was classified as “less than 10 years”, “11–15 years”, CHD deaths were observed in the group with non-SDB, “16–20 years” or “more than 20 years”. Diabetes was and 251 and 86 in the group with SDB, respectively. There considered present if the participant was taking insulin was a positive linear association between NC quartiles and or an oral hypoglycaemic agent. Hypertension was de- CHF incidence in the group without SDB (P =0.027), fined as systolic blood pressure ≥ 140 or diastolic blood while no association was observed in the SDB group. The pressure ≥ 90 mmHg or current use of antihypertensive same relationship also existed between NC and CHD medications [16]. Other covariates obtained from the par- mortality in the group without SDB (P =0.003) ent cohorts were age, sex, race, BMI, and levels of serum (Additional file 4: Figure S2). Survival analysis also re- triglycerides and high-density lipoprotein cholesterol. vealed the difference in the outcomes between these two subgroups. It was found that the CHD death-free curve in Statistical analysis the group without SDB had greater separation and more Data is presented as mean ± standard deviation for con- obvious consistency with NC quartiles than the one in the tinuous variables and number (percentage) for categor- SDB group. In addition, the tendency for a higher inci- ical variables. Cox proportional hazards models were dence of CHF with increasing NC quartiles existed in the used to evaluate the association between NC and inci- group without SDB only (Fig. 1). dent CHF or CHD mortality. The overall significance of In the group without SDB, after adjustment for age, sex, the association of NC with each outcome was tested and BMI, the HR for the incidence of CHF was 2.265 with NC modelled as a continuous variable. Survival (95% CI 1.074–4.777) in the highest NC quartile com- time was defined as the time from baseline polysomno- pared with that in the lowest NC quartile. After further gram to the first CHF (or CHD death) event. Censoring adjustment for waist circumference and other variables, time was the time of last known disease-free status for the association was diminished and became not signifi- those without cardiovascular disease. All analyses were cant. However, the association of NC, as a continuous adjusted in the model for the following covariates: (1) variable, with CHF incidence remained significant with an age, sex, and BMI; (2) age, sex, BMI, and waist circum- adjusted HR of 1.082 (95%CI 1.003–1.166) per unit in- ference; (3) age, sex, BMI, waist circumference, AHI, crease in NC after even adjusting for all of the variables in smoking status, serum levels of total cholesterol, triglyc- the group without SDB (P = 0.040). In the SDB group, the erides and high-density lipoprotein, history of diabetes, association was not statistically significant in both the uni- and history of hypertension, which are common risk fac- variate and multivariate models (Table 1). tors for cardiovascular diseases. The relationship be- Unlike the relationship between CHF and NC quartile, tween different quartile levels of NC and the incidence the HR for the incidence of fatal CHD was 5.477 of CHF or CHD mortality is presented as hazard ratio (95% CI 1.422-21.090) in the highest NC quartile com- (HR) [95% confidence intervals (CI)]. Net reclassification pared with that in the lowest NC quartile after adjusting improvement (NRI) index and integrated discrimination for all of the variables in the group without SDB. In improvement (IDI) index were calculated to assess the addition, after adjustment for age, sex, and BMI, the HR incremental value of NC in predicting cardiovascular of NC, as a continuous variable, for CHD death was diseases. P values < 0.05 were considered to be statisti- 1.141 (95% CI 1.014–1.282) and 1.061 (0.979–1.150) in cally significant. Statistical analyses were performed with the groups with and without SDB, respectively. In the SPSS software (version 22.0, IBM Corp., Armonk, NY, model with additional adjustment for waist circumfer- USA) as well as with R (version 3.0.1, R Foundation for ence, the association between fatal CHD and NC was Statistical Computing, Vienna, Austria). significant only in the group without SDB (P = 0.022). With all the covariates adjusted in multivariate model 3, Results there was still a significant association between NC and A total of 4433 subjects without CHD or CHF at base- CHD death in those without SDB, with an adjusted HR line were followed up for a median of 10.9 years. It was of 1.134 (95% CI 1.001–1.284) per unit increase in NC a predominantly Caucasian population (86.6%) with an (Table 2). average age of 63 years (results not displayed). The To further evaluate whether NC has an incremental group without SDB was younger (60.93 ± 11.05 years) value in predicting the risk of CHF and CHD death, IDI and had more female patients (67.4%), lower BMI, and and NRI were calculated in our study (Table 3). Reclassi- less cardiovascular risks such as dyslipidaemia, abnor- fication statistics showed a significant improvement in mal blood pressure, and diabetic status than the group NRI index of 0.2017 (P = 0.0135), indicating that, with Zhang et al. BMC Cardiovascular Disorders (2018) 18:108 Page 4 of 8 Fig. 1 Adjusted Kaplan-Meier survival curves for NC quartiles, according to the AHI group and event type. All the results were adjusted for age, sex, BMI, waist circumference, AHI, smoking status, total cholesterol, triglycerides, high-density lipoprotein, history of diabetes, and historyof hypertension. CHF = congestive heart failure, CHD = coronary heart disease, AHI = apnoea-hypopnea index, BMI = body mass index, NC = neck circumference the addition of NC, the prediction power of the model Discussion adjusted for age, sex, BMI has been improved, and The SHHS study was conducted in a community-based the proportion of correct classification increased by population and included ethnically diverse participants 20.17%. The prediction for CHD death were also im- with AHI status. It has reported several valuable findings proved (NRI = 0.2655, P = 0.0460). At the meantime, on the association between SDB and hypertension, the IDI index was 0.0060 (P = 0.0447) in the model CHD, and CHF [17–19]. To the best of our knowledge, predicting CHD death in the non-SDB population, our study is the first prospective cohort study to evaluate indicating that an aggregate measure of sensitivity the association between NC and future CHF events and and specificity was superior in the model adjusted fatal CHD in a large cohort. This study indicated that forage,sex,BMI andNCcompared withthe one participants without SDB with higher NC may develop adjusted for age, sex and BMI only. However, both more CHF events and CHD deaths. NC might be an reclassification statistics showed no significant im- early risk factor and showed preclinical predictive value provement in the predicting value with the addition for CHD death and CHF events in the population with- of NC in the SDB population. out SDB. Zhang et al. BMC Cardiovascular Disorders (2018) 18:108 Page 5 of 8 Table 1 The association of incident CHF with neck circumference according to AHI categories Quartiles of neck circumference Q1 (low) Q2 Q3 Q4 (high) Overall tendency AHI < 5 Subjects, n 805 655 491 283 Events, n (%) 39 (4.8) 45 (6.9) 49 (10.0) 30 (10.6) Univariate Model 1 (Ref) 1.457 (0.949–2.238) 2.197 (1.443–3.346) 2.341 (1.455–3.769) 1.095 (1.054–1.137) Multivariate Model 1 1 (Ref) 0.997 (0.630–1.577) 1.738 (0.982–3.077) 2.265 (1.074–4.777) 1.111 (1.034–1.194) Multivariate Model 2 1 (Ref) 0.891 (0.561–1.415) 1.386 (0.780–2.463) 1.762 (0.850–3.650) 1.080 (1.008–1.157) Multivariate Model 3 1 (Ref) 0.844 (0.522–1.364) 1.419 (0.778–2.589) 1.668 (0.765–3.638) 1.082 (1.003–1.166) AHI ≥ 5 Subjects, n 304 455 681 759 Events, n (%) 31(10.2) 51(11.2) 83(12.2) 86(11.3) Univariate Model 1 (Ref) 1.086 (0.695–1.697) 1.196 (0.792–1.807) 1.107 (0.734–1.669) 1.004 (0.974–1.035) Multivariate Model 1 1 (Ref) 1.051 (0.662–1.670) 1.065 (0.646–1.757) 1.107 (0.600–2.041) 0.999 (0.948–1.054) Multivariate Model 2 1 (Ref) 0.972 (0.609–1.550) 0.969 (0.587–1.599) 0.995 (0.544–1.820) 0.996 (0.947–1.049) Multivariate Model 3 1 (Ref) 0.870 (0.533–1.418) 0.807 (0.480–1.357) 0.690 (0.364–1.309) 0.965 (0.912–1.022) Model 1 adjusted for age, gender, and BMI Model 2 adjusted for age, gender, BMI, and waist circumference Model 3 adjusted for age, gender, BMI, waist circumference, AHI, smoking status, total cholesterol, triglycerides, high-density lipoprotein, history of diabetes, and history of hypertension Results are presented as hazard ratio (95% confidence interval). CHF congestive heart failure, AHI apnoea-hypopnea index (the average number of episodes of apnoea and/or hypopnea per hour of sleep), BMI body mass index P value for the overall tendency of neck circumference modelled as a continuous variable was <0.001, 0.004, 0.029, 0.040 in the univariate model, multivariate model 1, model 2, and model 3, respectively *P value for the overall tendency of neck circumference modelled as a continuous variable was 0.809, 0.978, 0.893, and 0.225 in univariate model, multivariate model 1, model 2, and model 3, respectively Table 2 The association of CHD death with neck circumference according to AHI categories Quartiles of neck circumference Q1 (low) Q2 Q3 Q4 (high) The overall tendency AHI < 5 Subjects, n 805 655 491 283 Events, n (%) 10 (1.2) 17 (2.6) 17 (3.5) 14 (4.9) Univariate Model 1 (Ref) 2.120 (0.971–4.631) 2.951 (1.351–6.445) 4.164 (1.849–9.374) 1.117 (1.049–1.189) Multivariate Model 1 1 (Ref) 1.622 (0.710–3.705) 2.760 (0.977–7.798) 5.480 (1.574–19.075) 1.141 (1.014–1.282) Multivariate Model 2 1 (Ref) 1.629 (0.714–3.717) 2.782 (0.992–7.807) 5.486 (1.624–18.527) 1.139 (1.019–1.274) Multivariate Model 3 1 (Ref) 1.914 (0.762–4.807) 3.427 (1.806–10.814) 5.477 (1.422–21.090) 1.134 (1.001–1.284) AHI ≥ 5 Subjects, n 304 455 681 759 Events, n (%) 8 (2.6) 18 (4.0) 34 (5.0) 26 (3.4) Univariate Model 1 (Ref) 1.479 (0.643–3.402) 1.886 (0.873–4.074) 1.291 (0.584–2.851) 1.013 (0.962–1.068) Multivariate Model 1 1 (Ref) 1.773 (0.749–4.194) 2.115 (0.858–5.215) 1.972 (0.663–5.868) 1.061 (0.979–1.150) Multivariate Model 2 1 (Ref) 1.643 (0.690–3.914) 1.891 (0.762–4.690) 1.735 (0.586–5.135) 1.050 (0.968–1.139) Multivariate Model 3 1 (Ref) 1.292 (0.504–3.309) 1.538 (0.579–4.087) 1.393 (0.472–4.538) 1.033 (0.944–1.130) Model 1 adjusted for age, gender, and BMI Model 2 adjusted for age, gender, BMI, and waist circumference Model 3 adjusted for age, gender, BMI, waist circumference, AHI, smoking status, total cholesterol, triglycerides, high-density lipoprotein, history of diabetes, and history of hypertension Results are presented as hazard ratio (95% confidence interval). CHD coronary heart disease, AHI apnoea-hypopnea index, BMI body mass index P value for the overall tendency of neck circumference modelled as a continuous variable was 0.001, 0.028, 0.022, and 0.048 in univariate model, multivariate model 1, model 2, and model 3, respectively *P value for the overall tendency of neck circumference modelled as a continuous variable was 0.616, 0.149, 0.236, and 0.478 in univariate model, multivariate model 1, model 2, and model 3, respectively Zhang et al. BMC Cardiovascular Disorders (2018) 18:108 Page 6 of 8 Table 3 Incremental value of neck circumference in predicting their secondary analysis, while another study found that the risk of incident CHF and CHD death NC independently contributed to the prediction of CHF CHD death cardio-metabolic risks that may be superior to waist cir- cumference or BMI [29]. Our study has also shown that AHI < 5 NRI 0.2017 (0.0420–0.3614) 0.2655 (0.0051–0.5260) a higher NC may contribute to CHF as well as CHD P 0.0135 0.0460 mortality, but only in the non-SDB population, after IDI 0.0055 (−0.0008–0.0118) 0.0060 (0.0001–0.0119) adjusting for BMI and other cardiovascular risk factors. P 0.0885 0.0447 The potential mechanism may involve elevated levels AHI ≥ 5 NRI 0.0081 (−0.1233–0.1395) 0.0862 (−0.1288–0.3012) of plasma FFAs [30]. The FFAs in circulation could in- P 0.9041 0.4331 hibit oxidation and glucose uptake, and impair insulin signalling transduction through a variety of pathways, IDI 0.0003 (−0.0005–0.0012) −0.0006 (− 0.0031–0.0019) leading to insulin resistance. In addition, FFAs are also P 0.3785 0.6398 associated with lipid metabolism disorder, vascular endo- Reclassification indices were calculated for the addition of neck circumference thelial injury, and hypertension [31]. Upper body SAT, in the model adjusted for age, sex, and BMI AHI apnoea-hypopnea index, BMI body mass index, CHF congestive heart typically represented by NC, has been demonstrated to failure, CHD coronary heart disease, NRI net reclassification improvement, be associated with a much larger proportion of systemic IDI integrated discrimination improvement FFAs [32]. Therefore, NC might be a much earlier signal Body fat is mainly located beneath the skin, around of predicting cardiovascular disease (such as CHF or the abdominal organs (stomach, liver, intestines, kidneys, CHD death) amongst the anthropometric indicators. etc.), within the bone marrow, and within muscles [20]. Additionally, as an index of upper body obesity, NC is It has been reported that nearly all of the major cardio- always recognized as a simple and time saving measure- vascular risk factors, such as blood glucose level, plasma ment. Compared with waist circumference, NC is less af- lipid ratio, and blood pressure worsen with obesity [21]. fected by the changes in the body size caused by Conversely, low BMI has also been shown to be a risk breathing, exercises, diet, and lifestyle habits such as factor for cardiovascular and bleeding events in a pro- drinking alcohol [33, 34]. The measurement of NC spective observational study [22]. Therefore, BMI might might be especially useful in special populations such as be not capable of reflecting the characteristics of local morbidly obese people, patients on bed rest, and preg- fat deposition as different compartments of body fat nant women [34]. have been demonstrated to be associated with heteroge- It should be noted that the predictive value of NC has neous physiological and pathological metabolism [23]. not been found in individuals with AHI ≥ 5 in our study, VAT accounts for 10–20% of the total body fat in men although the baseline characteristics showed that the and 5–10% in women, and has been reported to predict members of this group were older and suffered from cardiovascular diseases independent of traditional mea- higher BMI, NC, waist circumference, and other cardio- sures of obesity [24]. SAT, which is over 80% of the total vascular risks. It is proposed that the predictive value of body fat around the abdominal organs, buttocks, thighs, NC level in the SDB group may be diminished by the and neck, has drawn more attention in recent decades. strong association of SDB with NC and cardiovascular These subcutaneous layers are functionally distinct and diseases. First, studies found that NC was significantly independently correlate with metabolic complications higher in patients with severe SDB than in those with [25]. For example, gluteofemoral SAT was found to con- non-severe SDB whether obese or not, possibly due to tribute to the long-term entrapment of excess fatty acids the increased mass of upper respiratory tract soft tissues and protect from the adverse effects associated with ec- [35, 36]. In addition, SDB and NC were demonstrated to topic fat deposition [26]. Additionally, several studies be associated with lower levels of high-density lipopro- have reported that waist-to-hip ratio was strongly posi- tein and incidence of diabetes, which may directly con- tively associated with cardiovascular outcomes in com- tribute to cardiovascular diseases [34, 37, 38]. SDB may parison with BMI [27, 28]. also cause cardiovascular diseases by sympathetic ner- NC, as an anatomically separate component of the vous system activation and systemic inflammation body, is a distinct fat depot that represents the upper resulting from intermittent hypoxemia [39]. Therefore, a body SAT. The Framingham Heart Study demonstrated larger cohort of SDB and non-SDB population may be that NC was a novel measurement for cardio-metabolic needed to further investigate the interactions between risk and associated with cardiovascular disease risk fac- multiple risk factors. tors, including blood pressure, triglycerides, and fasting This study has certain limitations that deserve discus- blood glucose [3]. However, there was no statistically sig- sion. Firstly, although NC is a simple and feasible an- nificant association between NC and incidence of car- thropometric measurement, there are other factors that diovascular disease in the entire population according to influence NC other than fat deposition. Further imaging Zhang et al. BMC Cardiovascular Disorders (2018) 18:108 Page 7 of 8 studies on quantification of SAT of the neck are required Availability of data and materials The datasets used and/or analyzed during the current study are available to validate our study findings. Secondly, the small num- from the corresponding author on reasonable request. ber of CHD death events in the people without SDB did not exceed the number of covariates by at least Authors’ contributions GW, JJZ, QG, LYP, JML, YG, BY and BJF had full access to all of the data in the ten-times, which made the multivariate model 3 less study and take responsibility for the integrity of the data and the accuracy of powerful. Thirdly, as in an observational analysis, re- the data analysis and they drafted the manuscript. JJZ and QG extracted sidual confounding by unmeasured variables, such as data and performed the statistical analyses. The Working Group (GW, JJZ, QG, LYP, JML, YG, BY, BJF) and all authors contributed to interpretation of data, diet or physical activity, cannot be excluded, despite revised the article critically for important intellectual content and approved multivariate adjustment for existing risk factors. the final version of the manuscript. Ethics approval and consent to participate Conclusions The protocol was approved by the Institutional Review Board of each participating institution (Boston University, Case Western Reserve University, A higher NC might predict the incidence of CHF and Johns Hopkins University, Missouri Breaks Research, Inc., New York University CHD deaths in a community-based population without Medical Center, University of Arizona, University of California at Davis, University SDB. However, precise examination of the relationship of Minnesota – Clinical and Translational Science Institute, University of Washington) and signed informed consents were provided by the subjects. between neck SAT on cardiovascular outcomes and the possible underlying mechanism is worth investigations Competing interests in the future. The authors declare that they have no competing interests. Publisher’sNote Additional files Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Additional file 1: Figure S1. Flow chart of the study sample. AHI = apnoea-hypopnea index. (TIF 1700 kb) Author details Department of Emergency Medicine, the Second Affiliated Hospital of Xi’an Additional file 2: Table S1. Characteristics of subjects between the Jiaotong University, Xi’an 710004, China. Department of Cardiology, the population with SDB and non-SDB. (DOC 18 kb) Second Affiliated Hospital of Xi’an Jiaotong University, Xi’an, China. Division Additional file 3: Table S2. Characteristics of subjects by neck of Cardiology, Department of Internal Medicine, Tongji Hospital, Tongji circumference quartiles. (DOC 59 kb) Medical College, Huazhong University of Science & Technology, Wuhan, Additional file 4: Figure S2. The relationship between NC and CHF or China. Clinical Research Centre, the First Affiliated Hospital of Xi’an Jiaotong CHD death. There was a positive linear association between NC quartiles University, Xi’an, China. Department of Emergency Medicine, Longhua and CHF incidence or CHD mortality in the group without SDB, while no Hospital, Shanghai University of Traditional Chinese Medicine, Shanghai, association was observed in the SDB group. The rate of outcome events China. was represented within each NC quartile according to AHI categories. AHI = apnoea-hypopnea index, CHF = congestive heart failure, Received: 19 February 2018 Accepted: 23 May 2018 CHD = coronary heart disease, NC = neck circumference. (TIF 2144 kb) References Abbreviations 1. Fox CS, Massaro JM, Hoffmann U, Pou KM, Maurovich-Horvat P, Liu CY, et al. AHI: Apnoea-hypopnea index; BMI: Body-mass index; CHD: Coronary heart Abdominal visceral and subcutaneous adipose tissue compartments: disease; CHF: Congestive heart failure; CI: Confidence interval; HR: Hazard association with metabolic risk factors in the Framingham heart study. ratio; IDI: Integrated discrimination improvement; NC: Neck circumference; Circulation. 2007;116(1):39–48. NRI: Net reclassification improvement; SAT: Subcutaneous adipose tissue; 2. Karpe F, Pinnick KE. Biology of upper-body and lower-body adipose tissue– SDB: Sleep-disordered breathing; SHHS: Sleep Heart Health Study; link to whole-body phenotypes. Nat Rev Endocrinol. 2015;11(2):90–100. VAT: Visceral adipose tissue 3. Preis SR, Massaro JM, Hoffmann U, D'Agostino RB Sr, Levy D, Robins SJ, et al. Neck circumference as a novel measure of cardiometabolic risk: the Framingham heart study. J Clin Endocrinol Metab. 2010;95(8):3701–10. Acknowledgements 4. Goodpaster BH, Krishnaswami S, Resnick H, Kelley DE, Haggerty C, Harris TB, We appreciate the Brigham and Women’s Hospital for sharing the datasets et al. Association between regional adipose tissue distribution and both of the Sleep Heart Health Study (SHHS). The SHHS acknowledges the type 2 diabetes and impaired glucose tolerance in elderly men and women. Atherosclerosis Risk in Communities Study, the Cardiovascular Health Study, Diabetes Care. 2003;26(2):372–9. the Framingham Heart Study, the Cornell/Mt. Sinai Worksite and 5. Neeland IJ, Turer AT, Ayers CR, Berry JD, Rohatgi A, Das SR, et al. Body fat Hypertension Studies, the Strong Heart Study, the Tucson Epidemiologic distribution and incident cardiovascular disease in obese adults. J Am Coll Study of Airways Obstructive Diseases, and the Tucson Health and Cardiol. 2015;65(19):2150–1. Environment Study for allowing their cohort members to be part of the 6. Caffo B, Diener-West M, Punjabi NM, Samet J. A novel approach to SHHS and for permitting data acquired by them to be used in the study. prediction of mild obstructive sleep disordered breathing in a population- SHHS is particularly grateful to the members of these cohorts who agreed to based sample: the Sleep Heart Health Study. Sleep. 2010;33(12):1641–8. participate in SHHS as well. SHHS further recognizes all of the investigators 7. Fitch KV, Stanley TL, Looby SE, Rope AM, Grinspoon SK. Relationship and staff who have contributed to its success. A list of SHHS investigators, between neck circumference and cardiometabolic parameters in HIV- staff, and their participating institutions is available on the SHHS website, infected and non-HIV-infected adults. Diabetes Care. 2011;34(4):1026–31. https://sleepdata.org/datasets/shhs. 8. Torriani M, Gill CM, Daley S, Oliveira AL, Azevedo DC, Bredella MA. Compartmental neck fat accumulation and its relation to cardiovascular risk Funding and metabolic syndrome. Am J Clin Nutr. 2014;100(5):1244–51. SHHS was supported by grants (U01HL53916, U01HL53931, U01HL53934, 9. Hingorjo MR, Qureshi MA, Mehdi A. Neck circumference as a useful marker U01HL53937, U01HL53938, U01HL53940, U01HL53941, U01HL64360) from of obesity: a comparison with body mass index and waist circumference. J the National Institutes of Health. Pak Med Assoc. 2012;62(1):36–40. Zhang et al. BMC Cardiovascular Disorders (2018) 18:108 Page 8 of 8 10. Saka M, Turker P, Ercan A, Kiziltan G, Bas M. Is neck circumference 32. Nielsen S, Guo Z, Johnson CM, Hensrud DD, Jensen MD. Splanchnic lipolysis measurement an indicator for abdominal obesity? A pilot study on Turkish in human obesity. J Clin Investig. 2004;113(11):1582–8. adults. Afr Health Sci. 2014;14(3):570–5. 33. Luo Y, Ma X, Shen Y, Xu Y, Xiong Q, Zhang X, et al. Neck circumference as 11. Quan SF, Howard BV, Iber C, Kiley JP, Nieto FJ, O'Connor GT, et al. The Sleep an effective measure for identifying cardio-metabolic syndrome: a Heart Health Study: design, rationale, and methods. Sleep. 1997;20(12):1077–85. comparison with waist circumference. Endocrine. 2017;55(3):822–30. 34. Joshipura K, Munoz-Torres F, Vergara J, Palacios C, Perez CM. Neck 12. Dean DA, Goldberger AL, Mueller R, Kim M, Rueschman M, Mobley D, et al. circumference may be a better alternative to standard anthropometric Scaling up scientific discovery in sleep medicine: the National Sleep measures. J Diabetes Res. 2016;2016:6058916. Research Resource. Sleep. 2016;39(5):1151–64. 35. Ahbab S, Ataoglu HE, Tuna M, Karasulu L, Cetin F, Temiz LU, et al. Neck 13. Cho NH, Oh TJ, Kim KM, Choi SH, Lee JH, Park KS, et al. Neck circumference circumference, metabolic syndrome and obstructive sleep apnea syndrome; and incidence of diabetes mellitus over 10 years in the Korean genome and evaluation of possible linkage. Med Sci Monit. 2013;19:111–7. epidemiology study (KoGES). Sci Rep. 2015;5:18565. 36. KawaguchiY,FukumotoS,Inaba M, Koyama H, ShojiT,ShojiS, et al. Different 14. Redline S, Sanders MH, Lind BK, Quan SF, Iber C, Gottlieb DJ, Bonekat WH, impacts of neck circumference and visceral obesity on the severity of obstructive et al. Methods for obtaining and analyzing unattended polysomnography sleep apnea syndrome. Obesity (Silver Spring, Md). 2011;19(2):276–82. data for a multicenter study. Sleep Heart Health Research Group. Sleep. 37. Nieto FJ, Young TB, Lind BK, Shahar E, Samet JM, Redline S, et al. 1998;21(7):759–67. Association of Sleep-Disordered Breathing, sleep apnea, and hypertension in 15. Johns MW. A new method for measuring daytime sleepiness: the Epworth a large community-based study. JAMA. 2000;283(14):1829. sleepiness scale. Sleep. 1991;14(6):540–5. 38. Seicean S, Kirchner HL, Gottlieb DJ, Punjabi NM, Resnick H, Sanders M, et al. 16. Gottlieb DJ, Redline S, Nieto FJ, Baldwin CM, Newman AB, Resnick HE, et al. Sleep-disordered breathing and impaired glucose metabolism in normal- Association of usual sleep duration with hypertension: the Sleep Heart weight and overweight/obese individuals: the Sleep Heart Health Study. Health Study. Sleep. 2006;29(8):1009–14. Diabetes Care. 2008;31(5):1001–6. 17. Gottlieb DJ, Yenokyan G, Newman AB, O'Connor GT, Punjabi NM, Quan SF, 39. Drager LF, Mcevoy RD, Barbe F, Lorenzifilho G, Redline S. Sleep apnea and et al. Prospective study of obstructive sleep apnea and incident coronary cardiovascular disease. Circulation. 2017;136(19):1840–50. heart disease and heart failure: the sleep heart health study. Circulation. 2010;122(4):352–60. 18. Chami HA, Resnick HE, Quan SF, Gottlieb DJ. Association of incident cardiovascular disease with progression of sleep-disordered breathing. Circulation. 2011;123(12):1280–6. 19. Haas DC, Foster GL, Nieto FJ, Redline S, Resnick HE, Robbins JA, et al. Age- dependent associations between sleep-disordered breathing and hypertension: importance of discriminating between systolic/diastolic hypertension and isolated systolic hypertension in the Sleep Heart Health Study. Circulation. 2005;111(5):614–21. 20. Shen W, Wang Z, Punyanita M, Lei J, Sinav A, Kral JG, et al. Adipose tissue quantification by imaging methods: a proposed classification. Obes Res. 2003;11(1):5–16. 21. Lavie CJ, De Schutter A, Milani RV. Healthy obese versus unhealthy lean: the obesity paradox. Nat Rev Endocrinol. 2015;11(1):55–62. 22. Mak KH, Bhatt DL, Shao M, Haffner SM, Hamm CW, Hankey GJ, et al. The influence of body mass index on mortality and bleeding among patients with or at high-risk of atherothrombotic disease. Eur Heart J. 2009;30(7):857–65. 23. Vanderburgh PM. Fat distribution: its physiological significance, health implications, and its adaptation to exercise training. Mil Med. 1992;157(4): 189–92. 24. Mahabadi AA, Massaro JM, Rosito GA, Levy D, Murabito JM, Wolf PA, et al. Association of pericardial fat, intrathoracic fat, and visceral abdominal fat with cardiovascular disease burden: the Framingham Heart Study. Eur Heart J. 2009;30(7):850–6. 25. Smith SR, Lovejoy JC, Greenway F, Ryan D, de Jonge L, de la Bretonne J, et al. Contributions of total body fat, abdominal subcutaneous adipose tissue compartments, and visceral adipose tissue to the metabolic complications of obesity. Metab Clin Exp. 2001;50(4):425–35. 26. Manolopoulos KN, Karpe F, Frayn KN. Gluteofemoral body fat as a determinant of metabolic health. Int J Obes. 2010;34(6):949–59. 27. Lapidus L, Bengtsson C, Larsson B, Pennert K, Rybo E, Sjostrom L. Distribution of adipose tissue and risk of cardiovascular disease and death: a 12 year follow up of participants in the population study of women in Gothenburg, Sweden. Br Med J (Clinical research ed). 1984;289(6454):1257–61. 28. Larsson B, Svardsudd K, Welin L, Wilhelmsen L, Bjorntorp P, Tibblin G. Abdominal adipose tissue distribution, obesity, and risk of cardiovascular disease and death: 13 year follow up of participants in the study of men born in 1913. Br Med J (Clin Res Ed). 1984;288(6428):1401–4. 29. Zhou JY, Ge H, Zhu MF, Wang LJ, Chen L, Tan YZ, et al. Neck circumference as an independent predictive contributor to cardio-metabolic syndrome. Cardiovasc Diabetol. 2013;12:76. 30. Santosa S, Jensen MD. Why are we shaped differently, and why does it matter? Am J Physiol Endocrinol Metab. 2008;295(3):E531–5. 31. Colussi G, Catena C, Dialti V, Pezzutto F, Mos L, Sechi LA. Fish meal supplementation and ambulatory blood pressure in patients with hypertension: relevance of baseline membrane fatty acid composition. Am J Hypertens. 2014;27(3):471–81. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Cardiovascular Disorders Springer Journals

The association of neck circumference with incident congestive heart failure and coronary heart disease mortality in a community-based population with or without sleep-disordered breathing

Free
8 pages
Loading next page...
 
/lp/springer_journal/the-association-of-neck-circumference-with-incident-congestive-heart-wIZKpTUCA4
Publisher
BioMed Central
Copyright
Copyright © 2018 by The Author(s).
Subject
Medicine & Public Health; Cardiology; Cardiac Surgery; Angiology; Blood Transfusion Medicine; Internal Medicine; Medicine/Public Health, general
eISSN
1471-2261
D.O.I.
10.1186/s12872-018-0846-9
Publisher site
See Article on Publisher Site

Abstract

Background: Neck circumference (NC), representing upper body subcutaneous adipose tissue, may be correlated with increased risk of overweight/obesity, obstructive sleep apnoea, and metabolic and cardiovascular disease. However, the relationship between NC and the incidence of congestive heart failure (CHF) or mortality due to coronary heart disease (CHD) in a community-based population with and without sleep-disordered breathing (SDB) has not yet been clarified. Methods: We performed a prospective study using the Sleep Heart Health Study (SHHS) cohort. Cox proportional hazards regression models were used to estimate the association of different levels of NC with CHF incidence or CHD mortality in 2234 individuals with SDB and 2199 without SDB, respectively. Results: After adjusting for age, sex, and body mass index (BMI), NC was significantly associated with CHF when comparing the highest NC quartile group with the lowest (hazard ratio, HR, 2.265, 95% confidence interval, CI, 1.074–4.777) in the non-SDB population. This association diminished after further adjustment for other risk factors, but remained statistically significant, with an adjusted HR of 1.082 (95% CI 1.003–1.166) per unit increase in NC. Additionally, after adjustment for age, sex, and BMI, NC was also shown to be remarkably associated with CHD mortality (HR 1.141, 95% CI 1.014–1.282) per unit increase in NC in the non-SDB population but not in the SDB population. After adjustment for all the covariates, there was a significant association between NC and CHD death in those without SDB, with an adjusted HR of 1.134 (95% CI 1.001–1.284) per unit increase in NC. Conclusions: NC may correlate with CHF incidence and CHD mortality in population without SDB. NC measurement may help risk stratification for cardiovascular diseases. Trial registration: NCT00005275, January 1994. Keywords: Neck circumference, Congestive heart failure, Coronary heart disease, Mortality, Sleep-disordered breathing * Correspondence: gang_wang@mail.xjtu.edu.cn Jingjing Zhang and Qi Guo contributed equally to this work. Department of Emergency Medicine, the Second Affiliated Hospital of Xi’an Jiaotong University, Xi’an 710004, China Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Zhang et al. BMC Cardiovascular Disorders (2018) 18:108 Page 2 of 8 Background study, 1230 from the Cardiovascular Health Study, 688 Obesity, a pathological condition with body mass index from Framingham Offspring Cohort, and 1971 from (BMI) over 30 kg/m , is associated with a series of meta- other studies. Of them, 600 subjects were excluded be- bolic risk factors [1]. However, the overall obesity in- cause of pre-existing CHD, heart failure, or other related cludes unique intrinsic characteristics of different fat cardiovascular events, 762 because of lack of follow-up depots and the pattern of body fat distribution may also data, and 9 because of lack of NC data, ultimately leav- contribute to the prognosis [2]. Recently, significant het- ing an analytical sample size of 4433. SDB has been pro- erogeneity was found between visceral adipose tissue posed as a risk factor for cardiovascular events; besides, (VAT) and subcutaneous adipose tissue (SAT) of the SHHS mainly focused on the relationship between SDB upper body. VAT is recognized as a pathogenic fat de- and cardiovascular outcomes. Therefore, our investiga- posit with increased risk for insulin resistance, type II tion was carried out in the following two groups: the diabetes mellitus, and atherosclerosis [3, 4]. It was found group with SDB and the group without SDB (Additional that while lower body SAT may reduce the risk for car- file 1: Figure S1). diovascular disease, abdominal SAT was not related [5]. Previous studies have demonstrated that upper body NC and SDB fat produced the majority of systemic free fatty acids NC was measured during health interviews at the begin- (FFA), which was associated with insulin resistance, in- ning of the observation period. Participants were advised creased very low-density lipoprotein, and endothelial to sit upright and look straight ahead. An inelastic tape dysfunction [3]. Neck circumference (NC), measured at was applied around the neck just below the laryngeal the level of the laryngeal prominence, represents upper prominence. The NC measurement was made perpen- body SAT and may be correlated with increased risk of dicular to the long axis of the neck with the tape contact- overweight/obesity, obstructive sleep apnoea, type II dia- ing the skin surface under acceptable pressure. Subjects betes, metabolic syndrome, and cardiovascular disease were divided into NC quartiles of the following ranges: [3, 4, 6–10]. However, the relationship between NC and ≤34.1 (reference), 34.1–37.0, 37.0–40.5, and > 40.5 cm the incidence of congestive heart failure (CHF) or mor- [13]. Home polysomnography was performed with the tality due to coronary heart disease (CHD) has not yet Compumedics P-series portable monitor (Abbotsford, been clarified. Previously, the Sleep Heart Health Study Victoria, Australia). Detailed information were collected (SHHS) was carried out in a geographically diverse, as described in a previously published work [14]. Apnoea community-based population to assess the difference in was defined as complete or almost complete cessation of cardiovascular outcomes among participants with differ- airflow lasting for at least 10 s, while hypopnea was de- ent sleep apnoea statuses [11]. Based on SHHS, the fined as a clearly discernible decrease in airflow, or chest present study aimed to investigate the role of NC in the or abdominal plethysmograph amplitude that lasted for at incidence of CHF or CHD mortality in populations with least 10 s. Both, apnoea and hypopnea, required an as- and without sleep-disordered breathing (SDB). sociated 4% or greater oxyhaemoglobin desaturation. AHI is the average number of episodes of apnoea and/ Methods or hypopnea per hour of sleep [15]. SDB was defined as Study design AHI ≥ 5 respiratory events per hour while non-SDB was Analyses were carried out in two subsets of data ob- defined as AHI < 5. tained from the participants of SHHS (ClinicalTrials.gov Identifier: NCT00005275). A total of 5804 participants aged 40 years and older underwent home polysomnogra- CHD death and CHF phy and completed a set of questionnaires on general Baseline CHD or CHF was defined to be present if the health and sleep habits since 1994 [12]. Cardiovascular participant responded positively to a standardized ques- disease outcomes were tracked until 2010. Detailed aims tionnaire before the polysomnogram monitoring, or re- and design of SHHS are described elsewhere [11]. The ported that they had undergone coronary bypass surgery protocol was approved by the Institutional Review Board or coronary angioplasty, or if the parent cohort had an of each participating institution and signed informed identified CHD event or CHF before the SHHS baseline. consents were provided by the subjects. The data were CHD death was defined as fatal CHD at any time be- accessed based on a signed agreement with the Brigham tween the baseline polysomnogram and the final and Women’s Hospital. follow-up between 2008 and 2011. Incident CHF was de- fined as the first occurrence of heart failure during this Participants period of follow-up. Ongoing surveillance for CHD A total of 5804 subjects from SHHS cohort included deaths and incident CHF events was carried out by the 1915 from the Atherosclerosis Risk in Communities parent cohorts according to cohort-specific protocols. Zhang et al. BMC Cardiovascular Disorders (2018) 18:108 Page 3 of 8 Other covariates with SDB (P < 0.05) (Additional file 2:Table S1).More Smoking status was classified as “never” (if the partici- baseline characteristics are shown in Additional file 3: pant reported lifetime smoking of fewer than 20 packs Table S2. of cigarettes), “former”,or “current”. Educational level During the follow-up period, 163 CHF events and 58 was classified as “less than 10 years”, “11–15 years”, CHD deaths were observed in the group with non-SDB, “16–20 years” or “more than 20 years”. Diabetes was and 251 and 86 in the group with SDB, respectively. There considered present if the participant was taking insulin was a positive linear association between NC quartiles and or an oral hypoglycaemic agent. Hypertension was de- CHF incidence in the group without SDB (P =0.027), fined as systolic blood pressure ≥ 140 or diastolic blood while no association was observed in the SDB group. The pressure ≥ 90 mmHg or current use of antihypertensive same relationship also existed between NC and CHD medications [16]. Other covariates obtained from the par- mortality in the group without SDB (P =0.003) ent cohorts were age, sex, race, BMI, and levels of serum (Additional file 4: Figure S2). Survival analysis also re- triglycerides and high-density lipoprotein cholesterol. vealed the difference in the outcomes between these two subgroups. It was found that the CHD death-free curve in Statistical analysis the group without SDB had greater separation and more Data is presented as mean ± standard deviation for con- obvious consistency with NC quartiles than the one in the tinuous variables and number (percentage) for categor- SDB group. In addition, the tendency for a higher inci- ical variables. Cox proportional hazards models were dence of CHF with increasing NC quartiles existed in the used to evaluate the association between NC and inci- group without SDB only (Fig. 1). dent CHF or CHD mortality. The overall significance of In the group without SDB, after adjustment for age, sex, the association of NC with each outcome was tested and BMI, the HR for the incidence of CHF was 2.265 with NC modelled as a continuous variable. Survival (95% CI 1.074–4.777) in the highest NC quartile com- time was defined as the time from baseline polysomno- pared with that in the lowest NC quartile. After further gram to the first CHF (or CHD death) event. Censoring adjustment for waist circumference and other variables, time was the time of last known disease-free status for the association was diminished and became not signifi- those without cardiovascular disease. All analyses were cant. However, the association of NC, as a continuous adjusted in the model for the following covariates: (1) variable, with CHF incidence remained significant with an age, sex, and BMI; (2) age, sex, BMI, and waist circum- adjusted HR of 1.082 (95%CI 1.003–1.166) per unit in- ference; (3) age, sex, BMI, waist circumference, AHI, crease in NC after even adjusting for all of the variables in smoking status, serum levels of total cholesterol, triglyc- the group without SDB (P = 0.040). In the SDB group, the erides and high-density lipoprotein, history of diabetes, association was not statistically significant in both the uni- and history of hypertension, which are common risk fac- variate and multivariate models (Table 1). tors for cardiovascular diseases. The relationship be- Unlike the relationship between CHF and NC quartile, tween different quartile levels of NC and the incidence the HR for the incidence of fatal CHD was 5.477 of CHF or CHD mortality is presented as hazard ratio (95% CI 1.422-21.090) in the highest NC quartile com- (HR) [95% confidence intervals (CI)]. Net reclassification pared with that in the lowest NC quartile after adjusting improvement (NRI) index and integrated discrimination for all of the variables in the group without SDB. In improvement (IDI) index were calculated to assess the addition, after adjustment for age, sex, and BMI, the HR incremental value of NC in predicting cardiovascular of NC, as a continuous variable, for CHD death was diseases. P values < 0.05 were considered to be statisti- 1.141 (95% CI 1.014–1.282) and 1.061 (0.979–1.150) in cally significant. Statistical analyses were performed with the groups with and without SDB, respectively. In the SPSS software (version 22.0, IBM Corp., Armonk, NY, model with additional adjustment for waist circumfer- USA) as well as with R (version 3.0.1, R Foundation for ence, the association between fatal CHD and NC was Statistical Computing, Vienna, Austria). significant only in the group without SDB (P = 0.022). With all the covariates adjusted in multivariate model 3, Results there was still a significant association between NC and A total of 4433 subjects without CHD or CHF at base- CHD death in those without SDB, with an adjusted HR line were followed up for a median of 10.9 years. It was of 1.134 (95% CI 1.001–1.284) per unit increase in NC a predominantly Caucasian population (86.6%) with an (Table 2). average age of 63 years (results not displayed). The To further evaluate whether NC has an incremental group without SDB was younger (60.93 ± 11.05 years) value in predicting the risk of CHF and CHD death, IDI and had more female patients (67.4%), lower BMI, and and NRI were calculated in our study (Table 3). Reclassi- less cardiovascular risks such as dyslipidaemia, abnor- fication statistics showed a significant improvement in mal blood pressure, and diabetic status than the group NRI index of 0.2017 (P = 0.0135), indicating that, with Zhang et al. BMC Cardiovascular Disorders (2018) 18:108 Page 4 of 8 Fig. 1 Adjusted Kaplan-Meier survival curves for NC quartiles, according to the AHI group and event type. All the results were adjusted for age, sex, BMI, waist circumference, AHI, smoking status, total cholesterol, triglycerides, high-density lipoprotein, history of diabetes, and historyof hypertension. CHF = congestive heart failure, CHD = coronary heart disease, AHI = apnoea-hypopnea index, BMI = body mass index, NC = neck circumference the addition of NC, the prediction power of the model Discussion adjusted for age, sex, BMI has been improved, and The SHHS study was conducted in a community-based the proportion of correct classification increased by population and included ethnically diverse participants 20.17%. The prediction for CHD death were also im- with AHI status. It has reported several valuable findings proved (NRI = 0.2655, P = 0.0460). At the meantime, on the association between SDB and hypertension, the IDI index was 0.0060 (P = 0.0447) in the model CHD, and CHF [17–19]. To the best of our knowledge, predicting CHD death in the non-SDB population, our study is the first prospective cohort study to evaluate indicating that an aggregate measure of sensitivity the association between NC and future CHF events and and specificity was superior in the model adjusted fatal CHD in a large cohort. This study indicated that forage,sex,BMI andNCcompared withthe one participants without SDB with higher NC may develop adjusted for age, sex and BMI only. However, both more CHF events and CHD deaths. NC might be an reclassification statistics showed no significant im- early risk factor and showed preclinical predictive value provement in the predicting value with the addition for CHD death and CHF events in the population with- of NC in the SDB population. out SDB. Zhang et al. BMC Cardiovascular Disorders (2018) 18:108 Page 5 of 8 Table 1 The association of incident CHF with neck circumference according to AHI categories Quartiles of neck circumference Q1 (low) Q2 Q3 Q4 (high) Overall tendency AHI < 5 Subjects, n 805 655 491 283 Events, n (%) 39 (4.8) 45 (6.9) 49 (10.0) 30 (10.6) Univariate Model 1 (Ref) 1.457 (0.949–2.238) 2.197 (1.443–3.346) 2.341 (1.455–3.769) 1.095 (1.054–1.137) Multivariate Model 1 1 (Ref) 0.997 (0.630–1.577) 1.738 (0.982–3.077) 2.265 (1.074–4.777) 1.111 (1.034–1.194) Multivariate Model 2 1 (Ref) 0.891 (0.561–1.415) 1.386 (0.780–2.463) 1.762 (0.850–3.650) 1.080 (1.008–1.157) Multivariate Model 3 1 (Ref) 0.844 (0.522–1.364) 1.419 (0.778–2.589) 1.668 (0.765–3.638) 1.082 (1.003–1.166) AHI ≥ 5 Subjects, n 304 455 681 759 Events, n (%) 31(10.2) 51(11.2) 83(12.2) 86(11.3) Univariate Model 1 (Ref) 1.086 (0.695–1.697) 1.196 (0.792–1.807) 1.107 (0.734–1.669) 1.004 (0.974–1.035) Multivariate Model 1 1 (Ref) 1.051 (0.662–1.670) 1.065 (0.646–1.757) 1.107 (0.600–2.041) 0.999 (0.948–1.054) Multivariate Model 2 1 (Ref) 0.972 (0.609–1.550) 0.969 (0.587–1.599) 0.995 (0.544–1.820) 0.996 (0.947–1.049) Multivariate Model 3 1 (Ref) 0.870 (0.533–1.418) 0.807 (0.480–1.357) 0.690 (0.364–1.309) 0.965 (0.912–1.022) Model 1 adjusted for age, gender, and BMI Model 2 adjusted for age, gender, BMI, and waist circumference Model 3 adjusted for age, gender, BMI, waist circumference, AHI, smoking status, total cholesterol, triglycerides, high-density lipoprotein, history of diabetes, and history of hypertension Results are presented as hazard ratio (95% confidence interval). CHF congestive heart failure, AHI apnoea-hypopnea index (the average number of episodes of apnoea and/or hypopnea per hour of sleep), BMI body mass index P value for the overall tendency of neck circumference modelled as a continuous variable was <0.001, 0.004, 0.029, 0.040 in the univariate model, multivariate model 1, model 2, and model 3, respectively *P value for the overall tendency of neck circumference modelled as a continuous variable was 0.809, 0.978, 0.893, and 0.225 in univariate model, multivariate model 1, model 2, and model 3, respectively Table 2 The association of CHD death with neck circumference according to AHI categories Quartiles of neck circumference Q1 (low) Q2 Q3 Q4 (high) The overall tendency AHI < 5 Subjects, n 805 655 491 283 Events, n (%) 10 (1.2) 17 (2.6) 17 (3.5) 14 (4.9) Univariate Model 1 (Ref) 2.120 (0.971–4.631) 2.951 (1.351–6.445) 4.164 (1.849–9.374) 1.117 (1.049–1.189) Multivariate Model 1 1 (Ref) 1.622 (0.710–3.705) 2.760 (0.977–7.798) 5.480 (1.574–19.075) 1.141 (1.014–1.282) Multivariate Model 2 1 (Ref) 1.629 (0.714–3.717) 2.782 (0.992–7.807) 5.486 (1.624–18.527) 1.139 (1.019–1.274) Multivariate Model 3 1 (Ref) 1.914 (0.762–4.807) 3.427 (1.806–10.814) 5.477 (1.422–21.090) 1.134 (1.001–1.284) AHI ≥ 5 Subjects, n 304 455 681 759 Events, n (%) 8 (2.6) 18 (4.0) 34 (5.0) 26 (3.4) Univariate Model 1 (Ref) 1.479 (0.643–3.402) 1.886 (0.873–4.074) 1.291 (0.584–2.851) 1.013 (0.962–1.068) Multivariate Model 1 1 (Ref) 1.773 (0.749–4.194) 2.115 (0.858–5.215) 1.972 (0.663–5.868) 1.061 (0.979–1.150) Multivariate Model 2 1 (Ref) 1.643 (0.690–3.914) 1.891 (0.762–4.690) 1.735 (0.586–5.135) 1.050 (0.968–1.139) Multivariate Model 3 1 (Ref) 1.292 (0.504–3.309) 1.538 (0.579–4.087) 1.393 (0.472–4.538) 1.033 (0.944–1.130) Model 1 adjusted for age, gender, and BMI Model 2 adjusted for age, gender, BMI, and waist circumference Model 3 adjusted for age, gender, BMI, waist circumference, AHI, smoking status, total cholesterol, triglycerides, high-density lipoprotein, history of diabetes, and history of hypertension Results are presented as hazard ratio (95% confidence interval). CHD coronary heart disease, AHI apnoea-hypopnea index, BMI body mass index P value for the overall tendency of neck circumference modelled as a continuous variable was 0.001, 0.028, 0.022, and 0.048 in univariate model, multivariate model 1, model 2, and model 3, respectively *P value for the overall tendency of neck circumference modelled as a continuous variable was 0.616, 0.149, 0.236, and 0.478 in univariate model, multivariate model 1, model 2, and model 3, respectively Zhang et al. BMC Cardiovascular Disorders (2018) 18:108 Page 6 of 8 Table 3 Incremental value of neck circumference in predicting their secondary analysis, while another study found that the risk of incident CHF and CHD death NC independently contributed to the prediction of CHF CHD death cardio-metabolic risks that may be superior to waist cir- cumference or BMI [29]. Our study has also shown that AHI < 5 NRI 0.2017 (0.0420–0.3614) 0.2655 (0.0051–0.5260) a higher NC may contribute to CHF as well as CHD P 0.0135 0.0460 mortality, but only in the non-SDB population, after IDI 0.0055 (−0.0008–0.0118) 0.0060 (0.0001–0.0119) adjusting for BMI and other cardiovascular risk factors. P 0.0885 0.0447 The potential mechanism may involve elevated levels AHI ≥ 5 NRI 0.0081 (−0.1233–0.1395) 0.0862 (−0.1288–0.3012) of plasma FFAs [30]. The FFAs in circulation could in- P 0.9041 0.4331 hibit oxidation and glucose uptake, and impair insulin signalling transduction through a variety of pathways, IDI 0.0003 (−0.0005–0.0012) −0.0006 (− 0.0031–0.0019) leading to insulin resistance. In addition, FFAs are also P 0.3785 0.6398 associated with lipid metabolism disorder, vascular endo- Reclassification indices were calculated for the addition of neck circumference thelial injury, and hypertension [31]. Upper body SAT, in the model adjusted for age, sex, and BMI AHI apnoea-hypopnea index, BMI body mass index, CHF congestive heart typically represented by NC, has been demonstrated to failure, CHD coronary heart disease, NRI net reclassification improvement, be associated with a much larger proportion of systemic IDI integrated discrimination improvement FFAs [32]. Therefore, NC might be a much earlier signal Body fat is mainly located beneath the skin, around of predicting cardiovascular disease (such as CHF or the abdominal organs (stomach, liver, intestines, kidneys, CHD death) amongst the anthropometric indicators. etc.), within the bone marrow, and within muscles [20]. Additionally, as an index of upper body obesity, NC is It has been reported that nearly all of the major cardio- always recognized as a simple and time saving measure- vascular risk factors, such as blood glucose level, plasma ment. Compared with waist circumference, NC is less af- lipid ratio, and blood pressure worsen with obesity [21]. fected by the changes in the body size caused by Conversely, low BMI has also been shown to be a risk breathing, exercises, diet, and lifestyle habits such as factor for cardiovascular and bleeding events in a pro- drinking alcohol [33, 34]. The measurement of NC spective observational study [22]. Therefore, BMI might might be especially useful in special populations such as be not capable of reflecting the characteristics of local morbidly obese people, patients on bed rest, and preg- fat deposition as different compartments of body fat nant women [34]. have been demonstrated to be associated with heteroge- It should be noted that the predictive value of NC has neous physiological and pathological metabolism [23]. not been found in individuals with AHI ≥ 5 in our study, VAT accounts for 10–20% of the total body fat in men although the baseline characteristics showed that the and 5–10% in women, and has been reported to predict members of this group were older and suffered from cardiovascular diseases independent of traditional mea- higher BMI, NC, waist circumference, and other cardio- sures of obesity [24]. SAT, which is over 80% of the total vascular risks. It is proposed that the predictive value of body fat around the abdominal organs, buttocks, thighs, NC level in the SDB group may be diminished by the and neck, has drawn more attention in recent decades. strong association of SDB with NC and cardiovascular These subcutaneous layers are functionally distinct and diseases. First, studies found that NC was significantly independently correlate with metabolic complications higher in patients with severe SDB than in those with [25]. For example, gluteofemoral SAT was found to con- non-severe SDB whether obese or not, possibly due to tribute to the long-term entrapment of excess fatty acids the increased mass of upper respiratory tract soft tissues and protect from the adverse effects associated with ec- [35, 36]. In addition, SDB and NC were demonstrated to topic fat deposition [26]. Additionally, several studies be associated with lower levels of high-density lipopro- have reported that waist-to-hip ratio was strongly posi- tein and incidence of diabetes, which may directly con- tively associated with cardiovascular outcomes in com- tribute to cardiovascular diseases [34, 37, 38]. SDB may parison with BMI [27, 28]. also cause cardiovascular diseases by sympathetic ner- NC, as an anatomically separate component of the vous system activation and systemic inflammation body, is a distinct fat depot that represents the upper resulting from intermittent hypoxemia [39]. Therefore, a body SAT. The Framingham Heart Study demonstrated larger cohort of SDB and non-SDB population may be that NC was a novel measurement for cardio-metabolic needed to further investigate the interactions between risk and associated with cardiovascular disease risk fac- multiple risk factors. tors, including blood pressure, triglycerides, and fasting This study has certain limitations that deserve discus- blood glucose [3]. However, there was no statistically sig- sion. Firstly, although NC is a simple and feasible an- nificant association between NC and incidence of car- thropometric measurement, there are other factors that diovascular disease in the entire population according to influence NC other than fat deposition. Further imaging Zhang et al. BMC Cardiovascular Disorders (2018) 18:108 Page 7 of 8 studies on quantification of SAT of the neck are required Availability of data and materials The datasets used and/or analyzed during the current study are available to validate our study findings. Secondly, the small num- from the corresponding author on reasonable request. ber of CHD death events in the people without SDB did not exceed the number of covariates by at least Authors’ contributions GW, JJZ, QG, LYP, JML, YG, BY and BJF had full access to all of the data in the ten-times, which made the multivariate model 3 less study and take responsibility for the integrity of the data and the accuracy of powerful. Thirdly, as in an observational analysis, re- the data analysis and they drafted the manuscript. JJZ and QG extracted sidual confounding by unmeasured variables, such as data and performed the statistical analyses. The Working Group (GW, JJZ, QG, LYP, JML, YG, BY, BJF) and all authors contributed to interpretation of data, diet or physical activity, cannot be excluded, despite revised the article critically for important intellectual content and approved multivariate adjustment for existing risk factors. the final version of the manuscript. Ethics approval and consent to participate Conclusions The protocol was approved by the Institutional Review Board of each participating institution (Boston University, Case Western Reserve University, A higher NC might predict the incidence of CHF and Johns Hopkins University, Missouri Breaks Research, Inc., New York University CHD deaths in a community-based population without Medical Center, University of Arizona, University of California at Davis, University SDB. However, precise examination of the relationship of Minnesota – Clinical and Translational Science Institute, University of Washington) and signed informed consents were provided by the subjects. between neck SAT on cardiovascular outcomes and the possible underlying mechanism is worth investigations Competing interests in the future. The authors declare that they have no competing interests. Publisher’sNote Additional files Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Additional file 1: Figure S1. Flow chart of the study sample. AHI = apnoea-hypopnea index. (TIF 1700 kb) Author details Department of Emergency Medicine, the Second Affiliated Hospital of Xi’an Additional file 2: Table S1. Characteristics of subjects between the Jiaotong University, Xi’an 710004, China. Department of Cardiology, the population with SDB and non-SDB. (DOC 18 kb) Second Affiliated Hospital of Xi’an Jiaotong University, Xi’an, China. Division Additional file 3: Table S2. Characteristics of subjects by neck of Cardiology, Department of Internal Medicine, Tongji Hospital, Tongji circumference quartiles. (DOC 59 kb) Medical College, Huazhong University of Science & Technology, Wuhan, Additional file 4: Figure S2. The relationship between NC and CHF or China. Clinical Research Centre, the First Affiliated Hospital of Xi’an Jiaotong CHD death. There was a positive linear association between NC quartiles University, Xi’an, China. Department of Emergency Medicine, Longhua and CHF incidence or CHD mortality in the group without SDB, while no Hospital, Shanghai University of Traditional Chinese Medicine, Shanghai, association was observed in the SDB group. The rate of outcome events China. was represented within each NC quartile according to AHI categories. AHI = apnoea-hypopnea index, CHF = congestive heart failure, Received: 19 February 2018 Accepted: 23 May 2018 CHD = coronary heart disease, NC = neck circumference. (TIF 2144 kb) References Abbreviations 1. Fox CS, Massaro JM, Hoffmann U, Pou KM, Maurovich-Horvat P, Liu CY, et al. AHI: Apnoea-hypopnea index; BMI: Body-mass index; CHD: Coronary heart Abdominal visceral and subcutaneous adipose tissue compartments: disease; CHF: Congestive heart failure; CI: Confidence interval; HR: Hazard association with metabolic risk factors in the Framingham heart study. ratio; IDI: Integrated discrimination improvement; NC: Neck circumference; Circulation. 2007;116(1):39–48. NRI: Net reclassification improvement; SAT: Subcutaneous adipose tissue; 2. Karpe F, Pinnick KE. Biology of upper-body and lower-body adipose tissue– SDB: Sleep-disordered breathing; SHHS: Sleep Heart Health Study; link to whole-body phenotypes. Nat Rev Endocrinol. 2015;11(2):90–100. VAT: Visceral adipose tissue 3. Preis SR, Massaro JM, Hoffmann U, D'Agostino RB Sr, Levy D, Robins SJ, et al. Neck circumference as a novel measure of cardiometabolic risk: the Framingham heart study. J Clin Endocrinol Metab. 2010;95(8):3701–10. Acknowledgements 4. Goodpaster BH, Krishnaswami S, Resnick H, Kelley DE, Haggerty C, Harris TB, We appreciate the Brigham and Women’s Hospital for sharing the datasets et al. Association between regional adipose tissue distribution and both of the Sleep Heart Health Study (SHHS). The SHHS acknowledges the type 2 diabetes and impaired glucose tolerance in elderly men and women. Atherosclerosis Risk in Communities Study, the Cardiovascular Health Study, Diabetes Care. 2003;26(2):372–9. the Framingham Heart Study, the Cornell/Mt. Sinai Worksite and 5. Neeland IJ, Turer AT, Ayers CR, Berry JD, Rohatgi A, Das SR, et al. Body fat Hypertension Studies, the Strong Heart Study, the Tucson Epidemiologic distribution and incident cardiovascular disease in obese adults. J Am Coll Study of Airways Obstructive Diseases, and the Tucson Health and Cardiol. 2015;65(19):2150–1. Environment Study for allowing their cohort members to be part of the 6. Caffo B, Diener-West M, Punjabi NM, Samet J. A novel approach to SHHS and for permitting data acquired by them to be used in the study. prediction of mild obstructive sleep disordered breathing in a population- SHHS is particularly grateful to the members of these cohorts who agreed to based sample: the Sleep Heart Health Study. Sleep. 2010;33(12):1641–8. participate in SHHS as well. SHHS further recognizes all of the investigators 7. Fitch KV, Stanley TL, Looby SE, Rope AM, Grinspoon SK. Relationship and staff who have contributed to its success. A list of SHHS investigators, between neck circumference and cardiometabolic parameters in HIV- staff, and their participating institutions is available on the SHHS website, infected and non-HIV-infected adults. Diabetes Care. 2011;34(4):1026–31. https://sleepdata.org/datasets/shhs. 8. Torriani M, Gill CM, Daley S, Oliveira AL, Azevedo DC, Bredella MA. Compartmental neck fat accumulation and its relation to cardiovascular risk Funding and metabolic syndrome. Am J Clin Nutr. 2014;100(5):1244–51. SHHS was supported by grants (U01HL53916, U01HL53931, U01HL53934, 9. Hingorjo MR, Qureshi MA, Mehdi A. Neck circumference as a useful marker U01HL53937, U01HL53938, U01HL53940, U01HL53941, U01HL64360) from of obesity: a comparison with body mass index and waist circumference. J the National Institutes of Health. Pak Med Assoc. 2012;62(1):36–40. Zhang et al. BMC Cardiovascular Disorders (2018) 18:108 Page 8 of 8 10. Saka M, Turker P, Ercan A, Kiziltan G, Bas M. Is neck circumference 32. Nielsen S, Guo Z, Johnson CM, Hensrud DD, Jensen MD. Splanchnic lipolysis measurement an indicator for abdominal obesity? A pilot study on Turkish in human obesity. J Clin Investig. 2004;113(11):1582–8. adults. Afr Health Sci. 2014;14(3):570–5. 33. Luo Y, Ma X, Shen Y, Xu Y, Xiong Q, Zhang X, et al. Neck circumference as 11. Quan SF, Howard BV, Iber C, Kiley JP, Nieto FJ, O'Connor GT, et al. The Sleep an effective measure for identifying cardio-metabolic syndrome: a Heart Health Study: design, rationale, and methods. Sleep. 1997;20(12):1077–85. comparison with waist circumference. Endocrine. 2017;55(3):822–30. 34. Joshipura K, Munoz-Torres F, Vergara J, Palacios C, Perez CM. Neck 12. Dean DA, Goldberger AL, Mueller R, Kim M, Rueschman M, Mobley D, et al. circumference may be a better alternative to standard anthropometric Scaling up scientific discovery in sleep medicine: the National Sleep measures. J Diabetes Res. 2016;2016:6058916. Research Resource. Sleep. 2016;39(5):1151–64. 35. Ahbab S, Ataoglu HE, Tuna M, Karasulu L, Cetin F, Temiz LU, et al. Neck 13. Cho NH, Oh TJ, Kim KM, Choi SH, Lee JH, Park KS, et al. Neck circumference circumference, metabolic syndrome and obstructive sleep apnea syndrome; and incidence of diabetes mellitus over 10 years in the Korean genome and evaluation of possible linkage. Med Sci Monit. 2013;19:111–7. epidemiology study (KoGES). Sci Rep. 2015;5:18565. 36. KawaguchiY,FukumotoS,Inaba M, Koyama H, ShojiT,ShojiS, et al. Different 14. Redline S, Sanders MH, Lind BK, Quan SF, Iber C, Gottlieb DJ, Bonekat WH, impacts of neck circumference and visceral obesity on the severity of obstructive et al. Methods for obtaining and analyzing unattended polysomnography sleep apnea syndrome. Obesity (Silver Spring, Md). 2011;19(2):276–82. data for a multicenter study. Sleep Heart Health Research Group. Sleep. 37. Nieto FJ, Young TB, Lind BK, Shahar E, Samet JM, Redline S, et al. 1998;21(7):759–67. Association of Sleep-Disordered Breathing, sleep apnea, and hypertension in 15. Johns MW. A new method for measuring daytime sleepiness: the Epworth a large community-based study. JAMA. 2000;283(14):1829. sleepiness scale. Sleep. 1991;14(6):540–5. 38. Seicean S, Kirchner HL, Gottlieb DJ, Punjabi NM, Resnick H, Sanders M, et al. 16. Gottlieb DJ, Redline S, Nieto FJ, Baldwin CM, Newman AB, Resnick HE, et al. Sleep-disordered breathing and impaired glucose metabolism in normal- Association of usual sleep duration with hypertension: the Sleep Heart weight and overweight/obese individuals: the Sleep Heart Health Study. Health Study. Sleep. 2006;29(8):1009–14. Diabetes Care. 2008;31(5):1001–6. 17. Gottlieb DJ, Yenokyan G, Newman AB, O'Connor GT, Punjabi NM, Quan SF, 39. Drager LF, Mcevoy RD, Barbe F, Lorenzifilho G, Redline S. Sleep apnea and et al. Prospective study of obstructive sleep apnea and incident coronary cardiovascular disease. Circulation. 2017;136(19):1840–50. heart disease and heart failure: the sleep heart health study. Circulation. 2010;122(4):352–60. 18. Chami HA, Resnick HE, Quan SF, Gottlieb DJ. Association of incident cardiovascular disease with progression of sleep-disordered breathing. Circulation. 2011;123(12):1280–6. 19. Haas DC, Foster GL, Nieto FJ, Redline S, Resnick HE, Robbins JA, et al. Age- dependent associations between sleep-disordered breathing and hypertension: importance of discriminating between systolic/diastolic hypertension and isolated systolic hypertension in the Sleep Heart Health Study. Circulation. 2005;111(5):614–21. 20. Shen W, Wang Z, Punyanita M, Lei J, Sinav A, Kral JG, et al. Adipose tissue quantification by imaging methods: a proposed classification. Obes Res. 2003;11(1):5–16. 21. Lavie CJ, De Schutter A, Milani RV. Healthy obese versus unhealthy lean: the obesity paradox. Nat Rev Endocrinol. 2015;11(1):55–62. 22. Mak KH, Bhatt DL, Shao M, Haffner SM, Hamm CW, Hankey GJ, et al. The influence of body mass index on mortality and bleeding among patients with or at high-risk of atherothrombotic disease. Eur Heart J. 2009;30(7):857–65. 23. Vanderburgh PM. Fat distribution: its physiological significance, health implications, and its adaptation to exercise training. Mil Med. 1992;157(4): 189–92. 24. Mahabadi AA, Massaro JM, Rosito GA, Levy D, Murabito JM, Wolf PA, et al. Association of pericardial fat, intrathoracic fat, and visceral abdominal fat with cardiovascular disease burden: the Framingham Heart Study. Eur Heart J. 2009;30(7):850–6. 25. Smith SR, Lovejoy JC, Greenway F, Ryan D, de Jonge L, de la Bretonne J, et al. Contributions of total body fat, abdominal subcutaneous adipose tissue compartments, and visceral adipose tissue to the metabolic complications of obesity. Metab Clin Exp. 2001;50(4):425–35. 26. Manolopoulos KN, Karpe F, Frayn KN. Gluteofemoral body fat as a determinant of metabolic health. Int J Obes. 2010;34(6):949–59. 27. Lapidus L, Bengtsson C, Larsson B, Pennert K, Rybo E, Sjostrom L. Distribution of adipose tissue and risk of cardiovascular disease and death: a 12 year follow up of participants in the population study of women in Gothenburg, Sweden. Br Med J (Clinical research ed). 1984;289(6454):1257–61. 28. Larsson B, Svardsudd K, Welin L, Wilhelmsen L, Bjorntorp P, Tibblin G. Abdominal adipose tissue distribution, obesity, and risk of cardiovascular disease and death: 13 year follow up of participants in the study of men born in 1913. Br Med J (Clin Res Ed). 1984;288(6428):1401–4. 29. Zhou JY, Ge H, Zhu MF, Wang LJ, Chen L, Tan YZ, et al. Neck circumference as an independent predictive contributor to cardio-metabolic syndrome. Cardiovasc Diabetol. 2013;12:76. 30. Santosa S, Jensen MD. Why are we shaped differently, and why does it matter? Am J Physiol Endocrinol Metab. 2008;295(3):E531–5. 31. Colussi G, Catena C, Dialti V, Pezzutto F, Mos L, Sechi LA. Fish meal supplementation and ambulatory blood pressure in patients with hypertension: relevance of baseline membrane fatty acid composition. Am J Hypertens. 2014;27(3):471–81.

Journal

BMC Cardiovascular DisordersSpringer Journals

Published: May 31, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off