Antennal Journal of Cardiology 141 (2010) 122-138 ELSEVIER Contents lists available at ScienceDirect ICARDIOLoGy International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijeard Review The relationship of autonomic imbalance, heart rate variability and cardiovascular disease risk factors Julian F. Thayer **, Shelby S. Yamamoto °, Jos F. Brosschot © * the oan sae Unters. Deparment of Fhe 1635 Na Aven, Calumbus, hi 4210 US ® Mami sia of Pe Hea, Soa and Preveae Med. Mannheim Medea Fouy. Helelberg Univers dE ATeSr 7-1 -68157, Manheim Cormany Lien Unies, Dison of Chica and Heh Pcl. Departmen of scl Lien Unies, #0. Box 9555; 200 RB edn, The Needs ARTICLE INFO ‘ce ry Received 15 Jy 2009, Accepted 9 Septet 209 fsaaieomine 11 Rover 2000 Keywords ear rate variability Rk ators Hear eee Hypertesion ABSTRACT CGardiovasclar disease (CVD) isthe leading cause of death and disability worldwide. The understanding of the risk factors for CVD may yield important insights into the prevention, etiology, course, and treatment of ‘this major public health cancer, Autonomic imbalance, characterized by a hyperactive sympathetic system and a hypaactive parasympathetic system, is associated with various pathological condition. Overtime, excessive energy demands on the system can lead to premature aging and disease. Therefore, autonomic Imbalance may bea final common pathway to increased morbidity and mortality fram a ost of conditions nd disease, including etdiovascula disease, Heart ate varity (HRV) may be sed tasers autonomic imbalances, diseases and morality. Parasympathetc activity and HRV have been associated with 3 wide ‘ange of conditions including CVD. Here we review the evidence linking HRV to established and emerwing ‘modifiable and non-mesdifible CVD risk actors suchas hypertension obesity, family history and work tres Substantial evidence exists to suppor the notion that deceased HRV precedes the evelopment of amimber of risk factors and that lowering risk profiles i associated with increased HRV. We close witha suggestion tata ‘model of autonomic imbalance may provide a unifying framework within which o investigate the impact of Fis factors, including psychosocial actors and wark stress, on cardiovascular disease. 1. Introduction, Recent research has strongly suggested that negative affective states, dispositions and work stress are associated with diseases and i heatth [1-7]. Work stress, in particular, has been associated with substantial economic consequences, including increased absenteeism, increased worker turnover, decreased worker jab satisfaction and associated decreases in worker productivity [85]. tress at work is also a major public health risk associated with cardiovascular morbidity [10,11] Cardiovascular disease (CVD) is the leading cause of morbidity ‘and mortaliy in both men and women. This is paticularly tue in Seveloped countries [12,13] A wide range of risk factors for CVD have been identified but as yet a unified model that can account for this diversity of risk factors has not been put forward. ‘recent report from the Whitehall Study [11] has shown that, work stress associated with decreased heart rate variability (HRV). Decreased HRV is an independent risk factor for morbidity and ‘mortality. The important role thatthe vagus nerve playsin health and disease has been known for some time [14]. However, only relatively recently have researchers and clinicians started to investigate haw this * corresponding stor: Te Oia Sate Universi pre of acolo 1835. Nei Avenue Columbus, Oio4210.USA Tel: + 1614685 3450 fx + 614688825 Ema odes: Thayer 3980s (LF Thay) (0167-52738 ~ se front matter © 2009 Eee ela LA sas reserve "© 2009 EleevierIeand Le. Al sights reserved knowledge can be incorporated into a greater understanding of the ‘etiology, manifestations, course, outcomes, and treatment of disease, In this illustrative review we show that autonomic imbalance, in which vagal inhibitory influences are deficient, is associated with increased morbidity and all-cause mortality. We also review evi- ‘dence linking vagal function to established and emerging risk factors including work tres, for CVD and mortality. mportantly, we discuss ‘evidence that factors that increase HRV are associated with de- ‘reased risk and an improved health profile. Thus, the model of auto- ‘nomic imbalance may provide a unified approach to the understanding ‘of the re of HRY in the tsk for cardiovascular disease and all-cause morality, 2. Autonomic imbalance and disease “There is growing evidence forthe role ofthe autonomic nervous system (ANS) in a wide range of diseases. The ANS is generally con- ceived to have two major branches—the sympathetic system, asso- ciated with energy mobilization, and the parasympathetic system, associated with vegetative and restorative functions. Normally, the activity ofthese branches i in dynamic balance, However the activity ‘of the two branches can be rapidly modulated in response to changing ‘environmental demands. Conceptions of organism function based on ‘complexity theory hold that organism stability. adaptability, and health16 Taye neato onl of Cada 141 (2000) 12-131, ma are maintained through variability inthe dynamic relationship among, system elements [1,15-17]. Thus, patterns of organized variability, rather than static levels, are preserved in the face of constantly changing environmental demands. Because the system operates "far-from- cquilibrium,” the system is always searching for local energy minima ‘to minimize the energy requirements ofthe organism, ‘Another corollary ofthis view is that autonomic imbalance, where ‘one branch ofthe ANS dominates over the other, is associated witha lack ‘of dynamic flexibility and health Empirically, thee isa large body of evidence to suggest that autonomic imbalance, in which typically the sympathetic system is hyperactive and the parasympathetic system is hhypoactive, is associated with various pathological conditions [18]. In particular, when the sympathetic branch dominates for long periods of time, the energy demands on the system become excessive and ultimately cannot be met, eventuating in death. On the way to death, however, premature aging ad disease characterize asystem dominated by autonomic imbalance. Thus, autonomic imbalance may be a final ‘common pathway to increased morbidity and morality from a host of conditions and diseases, including cardiovascular disease. Heart rate variability cam be used to assess autonomic imbalances, diseases and mortality. Patasympathetic activity and! HRV have been associated with immune dysfunction and inflammation, which have been implicated ina wide range of conditions including CVD, diabetes, ‘osteoporosis, arthritis, Alzheimer's disease, periodontal disease, and certain types of cancers as well as declines in muscle strength and increased frailty and disability [3 19} Measures of heart rate variability (HRV) in both the time and frequency domains have been used successfully to index vagal activity In the time domain, the standard deviation of the interbeat intervals (IB), standard deviation of R to R intervals (SDNN), the root mean square successive differences (RMSSD), and measures of baroreflex sensitivity (an index of the responsiveness of the cardiovascular system to changes in blood pressure) have been shown to be useful indices of vagal activity. In the frequency domain both low frequency (LF: 0.04-0.15 Hz) and high frequency (HF: 0.15-0.40 Hz) spectral powers have been used as indices of vagal activity although there Is some debate over the branch of the autonomic system that affects these measures (20), Whereas there is little contention concerning HF power reflecting, primarily parasympathetic influences, LF power has been shown to reflect both sympathetic and parasympathetic influences. Neverthe- Jess, while there are some differences among studies, the consensus fs that lower values of these indices of vagal function are associated prospectively with death and disability. 3. Heart rate variability and mortal In one ofthe first studies to investigate the relationship between indices of HRV and mortality Kleiger etal. [21] showed in almost 900 ppost-myocardial infarction (MI) patients that HRV was a significant independent predictor of mortality in tis hgh risk group. Numerous stusies have since supported the notion that decreased vagal activity, as indexed by HRY, predicts mortality in high risk as well as low risk populations. nan elderly sub-sample of the Framingham Heart Study (EIS) frequency domain measures were significantly associated with all-cause mortality ater controlling for other risk factors. A total of 736 men and women with an average age of 72 years provided am- Dulatory time and frequency domain HRV data [22], Eight measures of FHV were examined including five frequency domain measures. All five frequency domain measures were significantly associated with all-cause mortality and all but the LF/HF ratio (2 putative measure ‘of sympathovagal balance where higher numbers indicate greater relative sympathetic dominance) remained so after controling for ‘other risk factors. A one standard deviation (SD) difference in the log ‘transformed LF power was associated with a 1.7 times greater relative risk of al-cause mortality in this sample [22 Similarly, in the Hoorn Study, a prospective study of glucose tolerance in the general popu- lation, several time and frequency domain indices of HRV were cal- culated and five were associated with all-cause mortality during the nine-year follow-up period at least at the p<0.10 level after con- trolling for age, gender, and glucose tolerance [23]. This finding was strongest for those at high risk because of diabetes, hypertension, or cardiovascular disease Inthe Atherosclerosis Risk in Communities (ARIC) study, the asso- ciation between HRV and mortality was vestigated in 11,654 men and women with an average age of 54 years [24]. Two minutes of supine resting beat-to-beat heart ate data were collected and a numberof time and frequency domain indices of HRV calculated. The lowest quartile of HE power was associated with incident MI, incident coronary heart disease (CHD), fatal CHD, and fatal non-CHD deaths in those with diabetes with hazard ratios ranging from 127 t02.03 over the eight-year follow-up petiod [24] In those individuals without diabetes the effects ‘were much ess consistent. However, examination of LF power indicated results consistent with the Framingham Study such that those non- diabetes in the lowest quartile had a 1.33 gteater risk of non-CHD ‘morality than those in the highest LF power quartile. The effect was even larger fr fatal CHD with those in the lowest LF power quartile hhaving a 1.92 greater risk than those in the highest quartile. In the ‘Autonomic Tone and Reflexes After Myocardial Infarction (ATRIMI) study, 1284 patients with a recent MI (within the last 28 days) were investigated using 24-h recordings (25].Forthe time domain measure of the SDNN and an abnormal score cu point of SDNN-=70 ms,a3.2 greater risk of mortality was found in the two-year follow-up perio. ‘Additionally, in a study of men and women post-MI with depressed left ventricular function (LVF), the HRV triangular index was used with an HRV cut point of -20 as an indicator of high risk [26}-In the placebo sroup, low HRV was a significant independent predictor of mortality With a relative risk of 1.46 after controlling forage, gender, LV ejection fraction, New York Heart Association class, diabetes, and beta-blocker tse. Thus, numerous studies have supported the notion that decreased vagal atvity, as indexed by HRY, predicts mortality in igh risk as well aslow tisk populations. However, an important caveat about al ofthese Stutles is that to date, few studies have examined the association between HRY indices and mortality in asymptomatic persons. 4, Heart rate variability and the etiology and progression of cardiovascular disease risk Theevidence for an association between reduced HRV and mortality, appears to be quite strong. Most ofthese studies examined the as- sociation after controlling for other known risk factors such as diabetes ane: hypertension. However, there is also evidence to suggest that reduced HRV leads to such risk factors, Thus, those studies that control for those known risk factors for which there exists evidence that reduced HRV might lead to those risk factors may, in fact, be under- estimating the role of vagal function in death and disease. ‘The National Hear, Lung, and Blood Institute ofthe US National, Institutes of Health lst eight risk factors for heart disease and stroke, six of which are modifiable. Three of these modifiable risk factors are associated with what could be called biological factors. They are high blood pressure (hypertension), diabetes, and abnormal cholesterol. ‘Three others listed as modifiable could be considered lifestyle factors and include tobacco use (smoking), physical inactivity (exercise), and obesity. The two non-modifiable isk factors are age and family history of early heart disease or stroke. Its interesting to note that there isat least some data to suggest that each ofthese risk factors is associated With decreased heart rate variability. 4.1. Modifiable biological risk factor: hypertension Perhaps the single most important risk factor for CVD is hyperten- sion, Numerous studies have documented the association between catdiac autonomic function and hypertension (Table 1). This associationmt {HE Mhayert nein ura of Carly 141 (2010) 122-131 ‘able ear at vais and hypectesin ties, ‘cork —_—stales ‘Subject and ample size Eectsnvestaated —_Contoed variates —_—Asacaton ‘yperenson Uso et ali] "HRV and yparesion Arce gender ‘Adjusted OF (O5EC): 1001.45 (051-348), furrent smoking dlabetes, 150 (055-350) and 244 (115-520) fom nd eduction | Iaghes to lowest quale of HV-HE and Incident typertension Hypertension Sigh etal.[28] _n=2024: 178 incident HRV and ypertemsion Age, BML smoking kobel_ Adjusted OR (95% CI: Men 138 (104-133, cs of hypertension censumption hase stole Women 1.12 (086-16) fr LF apd new-nsct and diastlc bled pressure hypertension Hypertension Sehocdereta [29] n=114061; 28% incident HRV. ypertension and_ Age. sea, ceter, Austed hazard rato (95% aes of hypertension Dowd pressure betes smoking. ‘stocton an EM SDNN 124 (110-140); RMSSD 136 (121-1358); irl 1.44 (127-163) frie ypeteasion ear ae variably igh quency. yas de, owen oe Toot mean sua of sucesivediferect. tral = cj eewoen two eonseNe R Wave, has been found in both cross-sectional and prospective analyses, Liao eta [27 | examined the association beeween two minutes of supine HRV and hypertension in a stratified random sample of 2061 blackand white ‘men and women from the ARIC study. During the three year follow-up period only 64 individuals developed hypertension. However, baseline HE power was inversely related to the development of hypertension among these individuals. ln cross-sectional analyses, HF power adjusted for age race, gender, smoking, diabetes. and education was significantly lower in the hypertensive group (both treated and untreated) than in the normotensive group. Additionally, those in the lowest HRV quartile hhad a 2-44fold greater risk of hypertension than those inthe highest quartile. In the Framingham Heart Study (FHS), Singh eta. [28] examined the association between two hours of ambulatory HR recordings and hypertension in men and women in cross-sectional and prospective analyses. Cross-sectional analyses indicated that after adjustment for age, BMI, smoking, and alcohol consumption several measures ofboth time and frequency domain indices of HRV were significantly lower in hypertensive men and women than in normotensives. During the four year follow-up period 119 men and 125 women developed hyper- tension. These analyses showed that low LF power was associated with the development of hypertension in men but notin women. Ina recent report the association between HRV, hypertension, and blood pressure was examined in 11,061 men and women ftom the ARIC study [29], HRV was assessed by 2-min and 6-min recordings separated by nine years. Consistent with previous reports HRV adjusted forage, race, study center, diabetes, smoking, education, and BMI was lower at baseline among those persons with hypertension. Importantly among, the 7099 persons without hypertension a baseline the lowest quartile of HRV as indexed by RMSSD adjusted for relevant covariates was associated with a hazard ratio forthe development of hypertension nine years later of 1.36 compared to those in the highest quartile These findings from large, epidemiological studies provide strong evidence that vagal tone, as measured by HRY, is lower in persons ‘with hypertension than in normotensives even after adjustment fora range of covariates, Importantly, these studies suggest that decreases in vagal tone may precede the development of thi critical risk factor for cardiovascular disease. 42. Modifiable biological risk factor: diabetes Diabetes, another important risk factor for CVD, has also been associated with decreased HRV (Table 2). Inthe first population-based study to examine the relationship between vagal tone, serum insulin, glucose, and diabetes, Liao etal. [30] investigated 154 diabetic and 1779 non-diabetic middle-aged men and women in the ARIC study. ‘Two minute supine resting HR recordings were used to compute HF ower as an index of vagal tone while fasting insulin, glucose, and ‘diagnosed diabetes were used to index diabetes and diabetes risk. Consistent with previous cross-sectional studies these researchers found that diabetics had lower vagal tone than non-diabetics after adjustment forage, race, and gender. In the nondiabetics, an inverse relationship was found between HF power, fasting insulin and fasting slucose, suggesting that reduced vagal tone may be involved in the pathogenesis of diabetes. However, after adjustment only the relationship between HRV and insulin. remained significant. This was the first study to examine the relationship between HRV and insulin and glucose ina general population and suggests that reduced ‘vagal tone may be involved in the pathogenesis of diabetes. Singh et al [31] examined the relationship between HRV and blood glucose levels in 1919 men and women from the FHS. The first two hours of ambulatory HR recordings were used to ealeulate a number of time and frequency domain indices of HRV. Fasting slucose levels were used to classify individuals as having normal or impaired fasting glucose, as well s toidentify those with diabetes (in addition to those with diabetic diagnosis). Several indices of HRV including LF and HE power in the FHS were inversely associated with fasting glucose levels and were significantly reduced in diabetics and those with impaired fasting glucose compared to those with normal fasting glucose levels [31]-The association between reduced HRV and diabetes remained significant after adjustment for age, gender, HR, BML, antihypertensive and cardiac medications, blood pressure, smoking, and alcohol and coffee consumption. Similarly, middle- ‘aged men and women fiom the ARIC study in the lowest LF power ‘quartile had a 1.2 greater fold risk of developing diabetes compared, to those in the highest quartile, after adjustment for age, race, ‘gender, study center, education, alcohol use, smoking, heart disease, physical activity, and BMI [32] Those with HR inthe highest quartile had 1.4 greater risk of diabetes than those in the lowest HR quartile with similar results for analyses restricted to those with normal fasting glucose, 43, Modifiable biological risk facto: cholesterol, To date few studies have examined the relationship between HRV. and cholesterol (Table 3), Of those that have studied this relationship,J Thayer ef | ternational oun of Caray 14 (2010) 122-131 ‘rate? Heart ate variability ad bees sus. CoD Tes Ses Subject and sample sae EME vested Conroe Vrs ‘Asotin Diabetes Tie ea 0) 1955; SE dabetcs HRV, diabetes, sting Age, race aeder “Adjusted gegnatse means ‘run ili and (bens. 078 or of dabets aucose hd 127 fr Hof non dlabetee (mean aifeencep=001): 134 (lowest quale) and 1.14 (highest quartile) offing serum isin and WF (p-001 “ortrend) for aed ad non -betes Dates Singh et [31] 110; 48 dabetcs HRV and blood Age. gender, HR BM Mea nL 64 for oral sting licose locos levels antihypertensive and cardi subjects and 6:54 for cabs melts ‘medications, loo presure, sublet (P= 0008) Smoking. an alcaholand eee Mean FHF: 1.2 for normal fasting shcose consumption subjects and 1.08 fr diabetes melts Sublets (p= 002) Diabetes Camethon eta [52] n= 8185: 13% dabetis HRVandType2 Age race. gender. study center Aste RR (95% Ch 12(1.0-14) fo saictes ‘tc, aleol use smoking. and 14 (12-17) for HR comparisons of beat dene prs nd MI ive the hest and hes quarties robb tow fequney igh equeney power ai evidence exists that low HRV is associated with high cholesterol levels. Christensen etal. [33] examined the association between 24- I HRV and cholesterol in 47 men with heart disease and 38 healthy ‘men. In both groups total cholesterol and low-density lipoprotein (LDL) were inversely associated with 24-h HRV. The association between HRV and cholesterol remained significant in both groups after adjustment for age and BM ‘The above results were also echoed ina study by Kupariet al. (34). Researchers investigating the association between short-term HRV and cholesterol among a random sample of 41 men and 47 women ‘without heart disease found that the RMSSD was inversely related {0 LDL cholesterol. Significance also remained after adjustment for ‘other potential confounders including physical activity, smoking, and alcohol consumption, 44, Modifiable lifestyle-related risk factors: smoking, physical inactivity, and being overweight Poor lifestyle choices. including a lack of physical activity and the abuse of tobacco, alcohol, and drugs have also been associated with autonomic imbalance and decreased parasympathetic activity [35-38] as wel as CVD. Of the modifiable lifestyl-telated risk factors for CVD (Table), pethaps the single most controllable rskis smoking. Hayano et al. [39] reported that both acute and chronic smoking was associated with decreased vagal tone. Likewise, smoking was associated with a significantly inereased LF/HE ratio within five minutesof exposure in taxi drivers under ordinary working conditions [40], Nighttime smoking in particular, appeared to have a more potent, acute effect on cardiac ‘modulation than daytime smoking. The authors also suggest that the sympathomimetricand parasympatho- withdrawal response ofsmoking ‘may have an adiitional roe in increasing cardiac risk (40). Ina very recent study, researchers have also found a link between. maternal smoking during pregnancy and heart rate variability among infants. Prenatal smoking was associated with lower RMSSD during quit sleep inthe first three days of ie [41], Minami al. 42] showed that indices of vagal tone inereased after ‘one week of smoking cessation in a group of habitual male smokers, Hayano etal [39] reported that both acute and chronic smoking was associated with decreased vagal tone. Importantly, Minami eta. [42] showed that indices of vagal tone increased after one week of smoking. cessation in a group of habitual male smokers. Moreover, ina study that examined the time course of the increase in vagal tone with smoking cessation Yotsukura et al [43] also reported that indices of vagal tone increased within 24 of smoking cessation. Interestingly, increases in vagal tone remained elevated for the one-month follow-up period in group of male habitual smokers. Thus, smoking and smoking cessation have imimediate and reversible effects on vagal ton ‘large number of cross-sectional as well as training studies have examined the effects of habitual exercise on cardiovascular Function, The single most replicable effect of aerobic training on cardiac function is a decreased resting HR. Whereas there is some ongoing debate about the nature of the autonomic nervous system changes that accompany regular physical activity numerous studies have implicated increased vagal tone in the salubrious effects of exercise [44]. We have reported in a cross-sectional study that habitual physical activity was associated with greater levels of vagally ‘mediated HRV in both men and women [37]. In 2334 men and 994 ‘women from the Whitehall study of British civil servants, moderate and vigorous physica activity was associated with greater vagal tone (in men) and lower resting HR (in men and women) compared to {those that reported low levels of physical activity after adjustment for age, smoking and alcohol consumption [45]. Another study among young adult also found that HF povrer rose ater 12 weeks of aerobic Conditioning among men but not women, which returned to pre- training levels after deconditioning [46]. Taken together numerous studies report that physical inactivity, an important lifestyle risk factor for CVD, is associated with decreased vagal tone. Importantly, i also appears that increased physical activity may decrease resting HR ‘and increase vagal tone. Studies have also dacumented reduced HRV among overweight, and obese individuals. In a study of 10 women with early-onset familial obesity and 10 non-obese women, several indices of HRV ‘were reduced in the obese women [47] Karason etal. [48 studied 28 ‘obese patients referred for gastroplasty, 24 obese patients using a lifestyle dietary modification approach, and 28 non-obese persons. At baseline both obese groups had reduced HF values relative tothe non- ‘obese participants After one year of follow-up, those persons that had undergone gastroplasty had an average weight loss of 28% and showed evidence of increased vagal function as indicated by increased HE power. Additionally, several studies of obesity in children and adolescence have also found that vagal function is reduced in obese individuals compared to non-obese individuals [49-51]. n all of these studies several indices of vagal function such as HF power were reduced in the obese individuals,1 JE Thayer ta terion ural of Carly 14 (2010) 122-131 ‘able Heart at vai and cholesterol. CDF OSes Suet ond Sale ae ees vee Conroe vaabes Aiton CGoleterol—Chtensn et a (33) =85; 554 wih heat ‘seas (en) Cholesteot Kupat eta 34 hort dase TRY adele 1=88; none cna with HRV and LDL cholesterol Physical activity. smoking, ends, ‘Mean SOR index ong i with THD dichotomized at oestera groups: 57 (low coker), 38 (high boestr}) ip-00s) Mean RNSSD among mea with IMD icone’ i cokestea 4s 5 (ow coker), 32 (high hoestr)}) ip-005) Mean SDNN index among eathy men Aihecomied it cokestea sans: 73 (low chokestera) 6 (high coester) (p-005) Mean RNSSD among heathy men Aihoronied it cholesterol 0s: 41 (aw colesterod) 32 (igh cokes) ( 456) 40 Pas “Diener 0M ae p99 Ure preMar nop] SIaYONeFEDAIWMaUL‘soL=u [6 RIHOHTYA SEINE HOM, 2H po ‘ae possi Burs ae sopuen De YoH EA gO" HON sspjzom au ang se1=u [oe] FP wOOMPUY HK SANE HON waka, jo pu sy urea R44 pu AH doo) fap Suse [c]aoeng pur ues] ‘SSUADEIAD YH noe ea mes s22n005 1) msi poy ss got sesame sua 661 = [acl e920) *(ouospus arog oto dost wes 0) 9 pep suorpuisoqeow pu aH} ssB}0m pes aE 991 =u lel wisauey — ssasa10q ua as uve ods pos ae (eo0=2) ov0~ ‘osné pur ng ens sages tour urous so 6 pe rw (S004) BO ASSAM BUETET _“no:teuonedena Yo] RUDRA pues eg aco pur ou gpg (gz) TewuauESI sas 10) ORE Syn pana paar AO 2a dus pur DOTS Sapa See GAO) ‘sous om pu AngeueN Ne UE“HE Thayer el ternational four of Carly 141 (200) 122-131 Py white-collar workers. One of the main results was that a one standard {deviation incteasein heart rate during sleep or ane standard deviation ‘decrease in nRMSSD was associated with significantly increased risk of mild hypertension, Similarly, 2 cohort study conducted among workers from the tegrated circuit manufacturing industry, waste incinerator plants and hospitals showed that various occupational factors were related to RY (80). Shift workers had significantly decreased SDNNi (mean of ‘the standard deviation of all NN intervals forall 5-min segments ofthe entire recording, in milliseconds), and increased LF and HR levels luring work compared to daytime workers. These workers also had significantly decreased SDNNi levels during sleep compared to daytime workers. Shift workers reporting acute high noise levels compared to low work noise levels also had elevated adjusted LF ‘means during work. Ths suggests that cardiovascular regulation is less successful among this group and could explain the excess cardiovas- cular disease risk among these workers. An additional finding related ‘to work stress was that significantly elevated LF means duting work adjusted for mean values during sleep were recorded among those in low job demand, low job control (77.9, 0.01), high demand, high job control (77.7, p=005) and high demand, low job control (7727, 1p <0.05) groups compared to a control group, after adjusting for sleep [80]. These results suggest that chronic disturbance ofthe autonomic cardiac balance favoring sympathetic dominance may be one reason {or the effects of workplace stress on CVD risk. The authors conclude that HRV can be a very useful tool to study work-related stress and their accompanying physiological effects Numerous studies have now reported that work tress s associated with increased risk of coronary heart disease (CHD) (11,81,82}. We also have previously shown that work-related worties were associated ‘with the largest increases in HR and decreases in HRV [83:4], Thus work-related stress as measured by job strain [2], effort-reward {mbalance | 1},and ecological momentary assessments [84]have been linked to decreased HRV. 6, Heart rate variability and the prevention of cardiovascular disease ‘There are several pathways via which the deleterious effects of modifiable factors such as work stress canbe prevented or minimized. All ofthese pathways involve efforts to increase HRV. As noted above, smoking cessation, physical exercise, and weightloss ae all associated with increased HRV. Dietary changes including the consumption of fruits and vegetables, moderate alcohol consumption, and intake of ‘omega-3 fatty acids and vitamin D through fish or nut consumption are also effective approaches for which there is some suggestive evidence linking them to increased HRV [85,86]. Stess and worry reduction via ‘meditation or worry postponement may provide effective ways to ‘prevent or minimize the effects of work stress [87]. Another possible approach as suggested by Tiller etal. [88], based on 24-h HRV re- cordings during normal working days. isthe use of postive emotions toaltersympathovagal balance. They suggest that this could be bene- ficial in terms of hypertension treatment and also reduce the risk of sudden death in those with congestive heart failure or coronary artery disease. Currently, however, there isa lack of studies investigating the {impact of such interventions on both HRV and disease. These types of studies could provide greater insight into the effects of autonomic ‘imbalance and new perspectives on the treatment and prevention of related diseases. 7. Summary Inthis paper we have tried to provide an overview of some of the evidence for the role of HRV in cardiovascular disease isk and ‘mortality. Although not exhaustive, this review shows that there isa large body of data to suggest that decreased vagal function is an independent risk factor for all-cause mortality. In addition, we exam- Ine evidence that decreased vagal function is a common factor inal of the majar risk factors for CVD, both modifiable and non-modifiable Furthermore, we showed that decreased vagal function characterizes emerging psychosocial risk factors. Importantly, the evidence pre- sented here also strongly suggests that work stress is associated with decreased HRV. Work stress itself Is a major risk factor for cardio- vascular morbidity and research suggests that one of the major pathways involves decreased HRV. We also note that the effects of ‘modifiable risk factors might be prevented or minimized by engaging in behaviors that might increase HRV, Moreover studies that include indices of both sympathetic and parasympathetic activity are nec- essary to provide a more complete pictute ofthe role ofthe ANS in health and disease (89). Finally, we suggested that autonomic im- balance might be the inal common pathway linking ahostof disorders and conditions to death and disease. Thus, behaviors that alter this ‘autonomic imbalance toward a more salubrious profile may serve to prevent or at least minimize the effects of certain factors on the risk for cardiovascular disease and death. ‘Acknowledgements We would lke to thank Tammy N.Sadle forher asistance with the tables. The authors of this manuscript have certified that they comply withthe Principles of Ethical Publishing in the International Journal of Cardiology [0] References 111 Friedman Thayer Astonomic bane revises panic anita eat ate ‘arab. | Payson Res 1996-4139 31 12) Kang MG Ko SB Cha 8, Park Wo JM, Chang. Assexiation bere ob ‘Sretson beat ate varity nd etabolsyadrome a shipyard male workers ‘Yonse Med J 2008 45:535-06 |p) Kiea-Caser Je MeGue L, Robles T, Chser R Eotons, mediiy, and Iota: pew perspectives fom psichoneuroimmunoloy. Aa Rev Psychol asst 83-107 1a) Aran 0s, Meceney MK_ Ect of psjcologial and socal factors on organic duce coed asteament of cesearch oa coronay Heart duce aon Be Psyc 200295 31-69. Ist Musselman DL, vane DL, Nemerofl CB The rationship of depression 10 ‘raivascub disease: epidemiology, logy, an eaten Arch Gen Sj Sty 1o0835580-92 U6 Rta A Bnet J Kanan |. opt of pec ats om the 17] Vere RL, Mien MA. The impact f emotions onthe hear. Prog Brain Res [suis SF ar Suen GME van den rand PA, Zeges MPA metaanays of servation sis Weis predictors sckess absence. lin Edema 20750:105-13. to] Hamer Smt FL Hays TL sins ani evel lationship cen emplsee Sacto. emplyee engagement an ass: eters et aa] APP Foyeol 20023 255-79 10] Rewengren A Haken §OunpetalAsociatonfpychosaa i actors with Fri of sete moar nation in 41119 exes ae 1308 contra fro 52 exis (the INTERIEART stone cnt sty. Lane 200964959 82 un) Chanda rit Bruner Ee Work tess and coronary hear dese: ‘wtarare the machansns? fur Heat | 200829:00- 8 [12] Rtutay Ch, toper AD The lobal burden disease: a compebensve sessment ‘mayan dt fom dee nies ar risk fact 1990 ad Prof to 2002 Boston, MA: Harvard School of Public Hel 1286, 13] Muar Rey Oungu Anand ctl burden ofeardvascubr dieses: pat: faneral eosin, the epdemilope walle, Fk aco and npc of {inten chalstion 201 1082765 114] Daryn The expression ofemtion in man apd animal. New York: Oxford University Pres 1958 115] Bred 6 Thayer Act ad anomie: a advasclaraprech Souja 19.30.25 11 Re ean Te at fae 3 yal sts apr 117] Thayer ane KD. A modelo neuroisceral integration in emotion regulation and dysregulation | Act Disord 2000;61:201-16 [19] Malan Pagan M, Lonard. Metbods for ssssment of sympsthosagal Balance: power spect analy. Aeon, New Yr: Futur: 18 1 Ener a Agen snd i ge xpos10 JE Thayer tab eternal our of Carey 141 (2010) 122-131 120] EkectrophsilonyTFeRESOCaNASOP Heat ate varity: standards of measure ‘xomsre tao dsp in the viscose istry. Ine Ach Occup Environ ‘net slope merpestton and cal use Crclaon 199683:104-05 ean 1977024836 [21] Keiger RE Miler, ger ef, Moss AL Deceased ea te varity apts 58] Pzas MY, oie M, Mass ft uece of gender ad andy history of ‘ssw with ineaed tay afer ae joer iron Ars] Cdl hypertension on sutonemie contol a heat ae, dase funtion sod beh iaysoae-c2 net pepe Hypertens 20018 143-8 [22] Tsu Veni jr Manders ES tal Reduced hear ratevabilty apd morality 35] Pree, VeaE, Neo M,DarateM, Tanti 8, Margo V. Autonomic ‘inaneldey cohort The Framingham Her Study. Creston 1994208788 Imodultion of here rat and bod pressure In normotensive cspring of [23] Gerstsen Deter) enone el Inparedauonomctncton sasscated Iypetense subjects. ab Cin Med 200: 135185-52 nth ncesed marty, especially i subjects with dae, hypertension, or [36] Naver Sic M Aces. Atoaomic eves ye an mova ations toy of cardvascar ease: the Hove sty Dates Cae 200128 1795-8 ‘mati haar of ypetenson lon Press 208 15-10 [24] tino, Cametion Mt sans Cw Cascio WE Hes Lowe eat rate vary 57] De Angels Perl rea Reta Modified autonomic balance negrings of socited hhe develpe teeny hea dese vis ith lates Slabs detected by Spectral arly afhea rateYaraliny.Netab st 2001 0: theatheraaerss st nears (AC) Su) Daetes 2025 324-3. ta [25] La Rovere Mr. Bigger Mares FL Motta Schwa Rarretex semsitivty 58] Lindi Wiklund, ec Ensim WV. Doestheatonomie nervous ten Se et at vary i prediction foal erdne moray ser nese play role tthe development sf insin easance? Asta om het te intron. ATRAWI (autonomic tone and refleks aer mec inrton) arab n fst degre reais Type 2 atte patients and ont ss Investigators tance 1958391478" Die Mod 20:20 90-405, [a5] Camm A at OM, Sekar Me al Moray in patents ater a recent 39] Hun MV. jaknen VSjuane Me a Her rte varity and pogesion of ‘myocar infrcon: 3 randomized. plcebe controled rl of amide Ung foroniyaersclerois Arter Thromb Vase Bil 19:1 1979-85 et ate vray for isk seatietion Coat 208109906 [so] Kenny HL Bta nce, vennclr ath and se cardi det, [27 lino D.C} Rares RW, et 3 scan of eric astennmse ncn 3 the fond cara 870-295 evelopment of hyperesin: he ARK sty. A] Hypertens 19963°1147-95. [61] Lampert R, Bremner JD Sse Deceased heart ate varity asi with (es) Sigh Larson MC, Tsuji sans JC. @Doonel Levy D. Reduced berate grees nara a lddle-aged nn An Het) 206158759017 ‘atlably and new-onset hypertension: iiss ito pathgenest of hyperen- 62] Sein PX Burlay JL Caves PM et al Higher evel of flanmaton factors ard Son the Naming Heat Stuy. Hypertension 1908/3225" {peaer ila rerstance are ndependenly esocated with higher eat td [29] Schroeder 8 0D, Chamblee Pinas Evans GW, Hes Hypertension, fewer ear rte vara in normoglyeeic elder ndvduals the Carioas- Dowd pes and heart ate variably the Atherosclerosis Rsk in ommunites ful Heath Sty] Arn Crat So 2008 35 315-21, (ac) sty. Hypertension 20054211061 ss] Barth Schumacher MU Hewmann-ingen © Depression asa rk fxr for [30] Lao, Cal ane eta. Asotin of al coe wit serum insulin, gcse, Iaraiy in patents wiih coronary heart disease: 2 metas Sjeosam ed dabetesmlitas the AL Stdy- Dates Res Ca Pt 195;30.21-21 ed 2ooseea03 12 [b1) Sigh Larsen MC. ODomellC.etak Assocation oFypeayera with reduced 18] Kubeansy LD, Kawachi Going the heart ofthe mater do negative eosons erate warty (he raingham Her Sud) A Cac 200085 308-12 {aus coronary heat disease? | Payson es 200048 323-37 [32] Gmethon hi ort SPFalantapan Duncan BX Sct ML. Chambles 65] Yan Melle, conse. spukerman TA eal Prognostic assortion of depression {Rak cto or popreston to incon! iyperaenemia: the Aerosceens Fellow moar irc with ml at cadionaseaar event eta ain Courin Stdy, 187- 1908: An} Eden 2002 138-108-67 nals ychosom Med 200¢66814-22. [33] Chstesen Ja To Chsensen MS. Schl EB Heat fate vata and 8) Sheps DS, SeleldD. Depression, ait, and the cahovaseular spstem: the asa lip in men with and without ischaemic heart disease Aterosteess Carlos’ perspective li Psy 20162 Suppl8):12-6ascuson 7-8, foovrtasisi- 67] Wort He Dopresien nd croracasr dase Cr Op Petry 30016221, [34] Kapaa MVrlien Koskinen. Tikkanen Mf Shor-crm hear rate varity] Yan LL Liu K Matthews KA, Daigls ML. Fetguson Tr. Kite C. sjchosaal set vg ik cry ade patna {es ni penn he Cac sk eel oon [251 Ingen aber CTyer J Redced heart ate waabity in chronic 6) asi’, Hawken S, Dna , ea. ct of potentially modi ik factors aleaol abs: rlatonstip with agai mond, cron hog suppressed sociated with myocar Iarcton i 52 counties (the INTESHEART 9) femmpulsve dnking Rel Pena 200354 1827-38 faseconol study Lancet 2004 368557-52 [a5] eed orges SW. Newlin Di. Elect of ol on vagal regulation a tio. [70] Everson ase SA, Hoare. Mero Re Depresine symptoms and mara ssn 2 ‘asc fonctn:contintons ofthe plyagl theory fo the pyebonhy sology atonal sample: confounding eet of health sats, Phos Med 200455 ‘falco. xp Clin Psychopharmacol T9987 84-82. 0, [37] Rosy LA Thayer Fes an gender-eteéaiferencesinhear period varabity. [71] Maced 1. Davey Smith G Heap Petal Carol 0, Har Psyc Peporam Mad 1n80.77-81 Stress and eadiovascuar seace: empirical demonsaton of Bus i a [bs] Weise rl, Bink Ace alohaingston ees ear ate varity prospective servation tly of cosh men Bh 20025241247-5 Drug Alao Depend 16:17 80-81 172] Thayer J Friedman BH, Dorowee TD. Alona characteris of aenrazed [b9] Hoysno Yama Me Saha Ya Short and ing-erm ees of carete anne isrder ae worry. ol Pthiy 1986 39.2556. Smoking on hear rate vanity. Am aol 198005388, 173] Cane RN Fealana KE Nile CE. Jil AS. Dees ik tor for aac [ao] Kabah F Waabe T Atmata Yet al Ace fs of caren smoking on he natty ad maida eve of pce meni coc Res 200 53 ert abt ders ding work Sal] Werk Env Heath 208 toro. oe [74] Cohen H, Maar MA Kaplan Z Kode M, Power spectral amiss of eat cate 11] Fler Wr, gers ST Youngman M, Gomez Griben EM NOM, Es fol ‘arab in psy. Psyche Pycosm 1999;65:38-66 fin snang during reayon at autonomic one Dev Pychate 200851: 75] Thayer, iedman BH. Aerevscalitgraon moet of health dispar in nea 2 Washington, DC: The Nasal academies Pes: 2008 [2] Mina. shmisu, Matsuoka Hees ofsmoing cessation on bond pressue 76) Ruse van Daoinen {Houtman De Gaus ob stl a raion co tie beat rate vara bial sates Hypertension 199.29 586-50, buoy blood press, heat ate a eat rate Yay ann lemale 1p] Yowura M Koide V. Fu Ket al Heart varabty during the est ronth oF uses Scand Work Enon Hel 2004 30:477-85, Smoking cessation At Hea] 198135 100-8 1771 Tsaneva Dako Cortetors been tain hearing changes nd vegetative [a4] Gite, anster EW, Baber AP Ect of endurance exercise on autonoic Balance miners Cent Eur Jub Health 2004 120-52 entol o hertrate. Sports Med 2003:335-45, [73] Heinen Moai M, Plone 5a Ello reward ible, beat et [45] Kenai Kl, Hema He Kamas M, Bruner E MGB ME, Marmot Mlle of fd heat te varity the Cadvasae tin Young Fas Sl. fj ‘aerate nd goo plist on eat te vary oa Bs tay chav Me 20s a:02"12 scl servants rn Epideil 2008 158:135-43. [p9]Vajkot Tv Boome de Ges fects of work tess on ambulatory ood 145] Shon RP, Shapiro PA, Beers Rt The effect of aerobic ening ad pressre, part ate and hear fate afabity Hypertension 200333806, {ard nono regulation a young ads Am Publ Heat 2009981-3. [80] Yan Aloo LG, Schouten, Maan AL Senne CA Ke F-Ocipatinal [a7] Hetretia a nade, de App Et al Asesament of cad atonomie termini of eat ate vat. lt Arch Occup Enron Heath 200073 entrlby heart pid arab iptens with eal onset late. a sre. Sehntoves nbs 36-32 1] AboaEhouleC Brisson Maunsell et a Job stan an ik of cute recurrent [a] Karson Molgard Wastrand .Sjastrom Heart rate vary n obey foronay heat disease even JAMA 200728: 152, nthe ec of weight sem Cal TED T2427 a2} Bele KL Landsberg PA Sena Laker Ds ob strain a major source of a9) va Para Baba Fe Obesity a atone action in aden, Carovaclresease tsk? Sean | Work ron He 2004:3085-128 in Bp Hypertens 200123 57-67. 1931 Brsscot J Van Dik f Thayer Je Daily worry related fo low Ret ete [50] Naga 8 Monta. Monta Asoc nwa sytem ety a he Varbity diving waking andthe subsequent nctrnal Sep pete. nt} Satean development oben pessoal cheren Obs es A511 25-52 Psychophys 207:85:0-47 [st] Sabbir. sie Contene cea Assesment fear autonome modulaton [84] Pipe’ rosso ean de eed Thayer. Carda ees of momentary suring adolescent obesty- Obes Res 2003:1:341-8. stead worry encode and ssl events Psychosom Med 20079: [52] Attn de Foula RS shnzato AR. eresCA Mansur A. Grpi uence fage, 185) Moca, iin Pines ScovikDS. Dear sal omega ty acd fended mas index and nsonal exp on en ate varsity 3 onsimption an es rte sara i Salts Caen 20081171130. {Shor of subjts witout art dese. Am] Carol 200495:381-5. [ao] Park Tock KL ONE MS ta. Prt vegetble and sh consumption and [ss] eiard Freer HBckel Aca Eectophytaloecal investigation of cal heat eva the Veterans Admiistauon Normative Ang Suey. A) peripberl ind anomie neve function te workers with lng level Clin wu 2oo.s0-78-86HE Thayer el eteratina ural of Carly 141 (2010) 122-131 nm [87] BrosschotyF Grin W-ThaerJE The perseveaie cognition hype 2revewof [89] Thayer JF Fisher JE Heat ate varity. overnight urinary norepineohne and wor. prolonged sessed ysl ctvaton al Peat] Pay osom Res (Create pote: evidence or the choline ainiamimatory palbway 1 Sogo 2 ely man lj tern Md 2000:265 439-17. [se] Tee WA McCay R Anson MCardiacoberence: ane, noninvasive measure [90] Costs Af Ee autor sd polishing lt Cadi 200:11:149-50, ‘tutor nervns spt de Alen er Heth dl 1986253-65