Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Prognostic value of heart rate variability in patients with coronary artery disease in the current treatment era

  • Antti O. Vuoti,

    Roles Formal analysis, Investigation, Writing – original draft

    Affiliation Research Unit of Internal Medicine, Medical Research Center Oulu, University of Oulu and Oulu University Hospital, Oulu, Finland

  • Mikko P. Tulppo,

    Roles Investigation, Writing – review & editing

    Affiliation Research Unit of Internal Medicine, Medical Research Center Oulu, University of Oulu and Oulu University Hospital, Oulu, Finland

  • Olavi H. Ukkola,

    Roles Investigation, Supervision, Writing – review & editing

    Affiliation Research Unit of Internal Medicine, Medical Research Center Oulu, University of Oulu and Oulu University Hospital, Oulu, Finland

  • M. Juhani Junttila,

    Roles Supervision, Writing – review & editing

    Affiliations Research Unit of Internal Medicine, Medical Research Center Oulu, University of Oulu and Oulu University Hospital, Oulu, Finland, Biocenter Oulu, University of Oulu, Oulu, Finland

  • Heikki V. Huikuri,

    Roles Conceptualization, Supervision, Writing – review & editing

    Affiliation Research Unit of Internal Medicine, Medical Research Center Oulu, University of Oulu and Oulu University Hospital, Oulu, Finland

  • Antti M. Kiviniemi,

    Roles Supervision, Writing – review & editing

    Affiliation Research Unit of Internal Medicine, Medical Research Center Oulu, University of Oulu and Oulu University Hospital, Oulu, Finland

  • Juha S. Perkiömäki

    Roles Conceptualization, Project administration, Writing – original draft

    juha.perkiomaki@oulu.fi

    Affiliation Research Unit of Internal Medicine, Medical Research Center Oulu, University of Oulu and Oulu University Hospital, Oulu, Finland

Prognostic value of heart rate variability in patients with coronary artery disease in the current treatment era

  • Antti O. Vuoti, 
  • Mikko P. Tulppo, 
  • Olavi H. Ukkola, 
  • M. Juhani Junttila, 
  • Heikki V. Huikuri, 
  • Antti M. Kiviniemi, 
  • Juha S. Perkiömäki
PLOS
x

Abstract

Coronary artery disease (CAD) mortality has declined substantially over the past decades thanks to advancing medical and interventional/surgical treatments; therefore, the prognostic value of the heart rate variability in CAD in the current treatment era is not well established. We evaluated the prognostic significance of baseline heart rate variability in 1,757 ARTEMIS study patients with angiographically verified CAD. During an average follow-up time of 8.7 ± 2.2 years, a total of 285 (16.2%) patients died. Of the patients, 63 (3.6%) suffered sudden cardiac death or were resuscitated from sudden cardiac arrest (SCD/SCA), 60 (3.4%) experienced non-sudden cardiac death (NSCD), and death attributable to non-cardiac causes (NCD) occurred in 162 (9.2%) patients. For every 10 ms decrease in standard deviation of normal to normal intervals the risk for SCD/SCA, NSCD and NCD increased significantly: HR 1.153 (95% CI 1.075–1.236, p<0.001), HR 1.187 (95% CI 1.102–1.278, p<0.001) and HR 1.080 (95% CI 1.037–1.125, p<0.001), respectively. The natural logarithm of the low-frequency component of the power spectrum and the short-term scaling exponent of the detrended fluctuation analysis also had significant association with all modes of death (p<0.001). After relevant adjustment, standard deviation of normal-to-normal intervals retained its association with NSCD and NCD (p<0.01), the natural logarithm of the low-frequency component of the power spectrum with all modes of death (p from <0.05 to <0.01), and the short-term scaling exponent of the detrended fluctuation analysis with SCD/SCA (p<0.05) and NCD (p<0.001). In conclusion, impairment of many measures of heart rate variability predicts mortality but is not associated with any specific mode of death in patients with stable CAD during the current treatment era, limiting the clinical applicability of heart rate variability to targeting therapy.

Introduction

Cardiovascular diseases still remain the leading cause of death in western societies and coronary heart disease (CHD) is the most common cause of death, accounting for approximately 40% of cardiovascular disease deaths [1,2]. Moreover, 80% of sudden cardiac deaths (SCD) occur in CHD patients and these patients are therefore at high risk of a sudden life-threating event [3]. Coronary artery disease (CAD) patients’ autonomic nervous system abnormalities can be evaluated non-invasively using different domains of heart rate variability (HRV).

The standard deviation of normal-to-normal intervals (SDNN) is a classic time-domain measurement of HRV. It reflects total variation of the heart rate signal, and reduced values of SDNN may indicate increased sympathetic activity or decreased parasympathetic activity [4] and have been associated with cardiac and other adverse outcomes [5,6]. High-frequency component of HRV power-spectrum (HF) reflects parasympathetic while low-frequency component (LF) reflects parasympathetic and sympathetic autonomic nervous control [7]. There are plenty of studies that have demonstrated the prognostic value of the frequency-domain measurements of HRV in different populations; for example, in one study in postinfarction patients, there was significant difference in LF component of HRV between those who died and survivors, whereas the difference in HF component was less marked [8]. The LF/HF ratio, a measurement of sympathovagal balance [9] and very low-frequency (VLF) component of HRV power spectrum have also been shown to yield prognostic information in postinfarction patients [10]. Some previous studies have suggested that some of the nonlinear HRV measurements may be slightly better predictors of mortality in postinfarction patients than the traditional measurements [1113]. The short-term scaling exponent of detrended fluctuation analysis (DFA1) describes the short-term fractal-like scaling properties of the RR interval time-series and is caused by complex interaction between the parasympathetic and sympathetic nervous system. Simultaneous sympathetic and vagal stimulation decreases DFA1 [14]. Reduced DFA1 has been shown to be associated with the risk of ventricular arrhythmias and death in various study populations [13,15]. The slope of the relationship between spectral power and frequency on bi-logarithmic scale shows a linear portion between 10−4 and 10−2 Hz (PL slope). The PL slope reflects the long-term scaling characteristics of HRV in the region of ultralow- and very-low-frequency bands [16]. A steep PL slope has been observed to be a better predictor of all-cause mortality or arrhythmic death than the conventional power spectral bands in postinfarction patients [11]. The data on the prognostic significance of approximate entropy (ApEn), a complexity measure of HRV, are limited. Decreased ApEn of HRV has been shown to precede spontaneous episodes of atrial fibrillation in patients without structural heart disease [17] and in patients after coronary artery bypass surgery [18].

In the modern treatment era, CHD mortality has declined substantially, and it has been estimated that around 45% of the decline is due to medical and interventional/surgical treatments [19,20]. However, the effect of this decline of mortality on the prognostic value of HRV in CAD patients is not clear. Furthermore, it is not well established whether decreased HRV is a general indicator of mortality risk or whether it is more closely associated with the risk of SCD. Therefore, the aim of the present study was to evaluate the value of conventional and nonlinear measurements of HRV in predicting mortality in patients with angiographically verified CAD in the modern treatment era.

Methods

Study population

The ARTEMIS study is a prospective observational study (Innovation to Reduce Cardiovascular Complication of Diabetes at the Intersection NCT01426685) that recruited patients between August 2007 and December 2012 at the Division of Cardiology in the University Hospital of Oulu [21,22]. The study included 1,946 patients with angiographically confirmed CAD defined as at least 50% stenosis in one or more major coronary vessels. Risk assessments were done 3 to 6 months after angiography. Of the study subjects, 833 had type 2 diabetes mellitus (DM), which was defined according to the WHO guidelines [23]. At the initial examination visit, laboratory analyses were done after overnight fasting using standard measures, and medical therapy for CAD and DM was optimized. Exclusion criteria were New York Heart Association class IV or Canadian Cardiovascular Society (CCS) class IV, planned or existing implantable cardioverter-defibrillator (ICD), permanent pacemaker, age under 18 or over 80 years, life expectancy <1 year due to any comorbidities, pregnancy, end-stage renal dysfunction requiring dialysis, or being otherwise unfit for the study due to physical or psychological condition. The study patients were contacted by mailed questionnaires and telephone calls to inquire about possible interim hospitalization at least at 2 and 5 years of follow-up. The final adjudication of the cause for hospitalization was based on diagnoses from medical records [21]. Due to CAD, patients were also under regular control by their doctors. Of the 1,946 patients, 189 did not undergo 24-hour ambulatory ECG or were excluded from the analysis as a result of technical and biological disturbances. Therefore, the total number of patients in the present study was 1,757, of whom 729 had DM. All subjects included in the study gave informed consent. The study was approved by the institutional Ethics Committee of Northern Ostrobothnia Hospital District and complied with the Declaration of Helsinki.

Endpoints of the present study

The endpoints of the present study were sudden cardiac death or resuscitated cardiac arrest (SCD/SCA), non-sudden cardiac death (NSCD), and non-cardiac death (NCD). The SCD/SCA group included patients who suffered SCD or were resuscitated from sudden cardiac arrest, whichever occurred first, and the criteria for SCD were witnessed death within 1 hour from the onset of symptoms or death occurring within 24 hours of last witnessed moment of being alive. NSCD included the patients who experienced death for cardiac reason but did not fulfill the criteria of SCD/SCA. The NCD group included all patients who suffered death due to other than cardiac reasons. The endpoints were determined from emergency rescue reports, hospital and physician reports, autopsy data, death certificates, and interviews with next of kin. The cause and mode of death were adjudicated by two independent investigators, and if needed, disagreement or uncertainty was resolved in consultation with the investigators (M.J.J. and H.V.H.) [21]. Autopsy data were available in most cases of SCD because according to the law, medicolegal autopsy is mandatory in Finland in cases of unexpected death.

Electrocardiography and heart rate variability

A digital Holter recorder (Medilog AR12; Huntleigh Healthcare Cardiff, U.K.) with an accuracy of 250 Hz was used to perform 24-hour ambulatory ECG recordings. During recording, patients were advised to carry on with their normal daily activities without limitations. The recordings were processed and analysis was conducted using HEARTS software (Heart Signal Oy, Oulu, Finland). Inter-beat intervals (R-R intervals) were edited visually using the interpolation method to remove all artefacts and ectopic beats from data. In this method, removed R-R intervals are replaced with a local average of the previous accepted normal R-R intervals [24]. In the present study, the following HRV parameters were included in the analysis as they represent widely different conventional and nonlinear domains of HRV and are most frequently used in clinical studies. Of the conventional HRV domains, SDNN was used as a time-domain measure and the included frequency-domain measures were the natural logarithm (ln) of HF power (ln(HF)) (0.15–0.4 Hz), ln of LF power (ln(LF)) (0.04–0.15 Hz), and ln of VLF power (ln(VLF)) (0.0033–0.04 Hz) of power spectrum and LF/HF ratio [9]. Nonlinear methods included DFA1, PL slope and ApEn. The detrended fluctuation analysis quantifies the presence or absence of the fractal-like correlation properties in RR-interval time series. In this method, root-mean-square is calculated from RR-interval time series which are integrated, divided to specific window sizes, and detrended. After repeating the steps mentioned above using different window sizes, HRV is plotted on a log-log scale as a function of the window size. DFA1 is the slope of this line in the window sizes ≤11 RR-intervals [25]. The PL slope method assesses the long-term fractal-like properties of the HRV [11]. ApEn is used to assess the complexity or regularity of the heart rate dynamics [26].

Echocardiography

Echocardiographic examinations were conducted using the General Vivid 7 ultrasound instrument (General Electric Healthcare, Little Chalfont, UK) to evaluate left ventricular ejection fraction (LVEF) and left ventricular mass index (LVMI). LVEF was determined using two-dimensional mode and LVMI was obtained by calculating left ventricle mass using the recommended formula and dividing it by body surface area [27].

Statistical analysis

Continuous variables are presented as mean ± standard deviation and categorical variables as percentages. The statistical significance of differences between baseline clinical characteristics was assessed using the standard two-tailed t-test for continuous variables and Chi-square test or Fisher’s exact test for dichotomous variables, as appropriate. The baseline clinical characteristics that differed significantly in univariate comparison were entered into the multivariate Cox regression analysis and the most predictive clinical model was defined individually for each endpoint group. HRV measurements that differed significantly in the baseline comparison were tested in the multivariate Cox regression analysis in the defined clinical model one at a time as continuous variables. Patient-specific follow-up times were individually applied for time-specific analyses due to varying follow-up times. A p-value < .05 was considered statistically significant. Association of HRV measurements with clinical risk indicators was assessed using the Pearson correlation coefficient between continuous variables and the two-tailed t-test between continuous and dichotomous variables. The receiver operating characteristics curves were used to optimize the cut-off points for HRV measurements. Kaplan-Meier curves were formed to show the cumulative proportional probabilities for different modes of death and Log rank test was used to assess the statistical significance of the separation of the curves. Analyses were conducted using IBM SPSS version 25 (IBM, Armonk, NY, USA).

Results

During a follow-up period of 8.7 ± 2.2 years, a total of 285 (16.2%) patients died in the study group of 1,757 patients. Of the patients, 63 (3.6%) suffered SCD/SCA and 60 (3.4%) NSCD. Death attributable to non-cardiac reasons occurred in 162 (9.2%) patients.

Baseline clinical characteristics of study patients

Many of the baseline clinical characteristics differed significantly between the patients who experienced a certain mode of death compared with the patients who remained alive/experienced a different mode of death (Table 1). The patients who suffered SCD/SCA were slightly older, had more commonly DM, were more often smokers, had more commonly CCS class ≥2, larger left ventricular end-systolic diameter, lower LVEF and higher LVMI compared with patients who remained alive/experienced other mode of death. Patients who died due to NSCD were older, had more commonly DM, consumed less alcohol, had more frequently CCS class ≥2, larger left ventricular end-systolic diameter and higher LVMI in comparison with subjects who remained alive/experienced other mode of death. Patients who experienced NCD were older, were more often males, had more commonly DM, were less often smokers, consumed less alcohol, had more commonly CCS class ≥2, had more often right bundle branch block and smaller left ventricular end-diastolic diameter compared with patients who remained alive/experienced other mode of death.

thumbnail
Table 1. Clinical characteristics and heart rate variability parameters in study patients at baseline.

https://doi.org/10.1371/journal.pone.0254107.t001

Association of baseline heart rate variability measurements with occurrence of different modes of death

The values of SDNN, ln(LF), ln(VLF), DFA1 and PL slope were lower in patients with each mode of death when compared with other patients. The patients who suffered NSCD or NCD had lower values of LF/HF ratio when compared with patients who remained alive/experienced other mode of death. The values of ln(HF) or ApEn did not differ significantly between patients who experienced any mode of death and other patients (Table 1).

Clinical models for different modes of death after adjustments

Clinical variables that differed significantly in univariate comparison (Table 1) were entered into the Cox multivariate regression analysis to define the most predictive clinical model individually for each endpoint group (Table 2). When age, DM, smoking, CCS class ≥2, left ventricular end-systolic diameter, LVEF and LVMI were tested in the Cox multivariate clinical model in a stepwise manner, age, DM, smoking, CCS class ≥2, LVEF and LVMI retained a significant association with SCD/SCA after adjustments. When age, DM, alcohol consumption, CCS class ≥2, left ventricular end-systolic diameter and LVMI were entered to the Cox multivariate clinical model in a stepwise manner, age, DM, CCS class ≥2 and left ventricular end-systolic diameter retained a significant association with NSCD. When age, gender, DM, smoking, alcohol consumption, CCS class ≥2, right bundle branch block, left ventricular end-diastolic diameter were tested in the Cox multivariate clinical model in a stepwise manner, age, gender, DM and left ventricular end-diastolic diameter remained significant predictors of NCD.

thumbnail
Table 2. Univariate and multivariate predictors of sudden cardiac death or sudden cardiac arrest, non-sudden cardiac death and non-cardiac death.

https://doi.org/10.1371/journal.pone.0254107.t002

Heart rate variability measurements in predicting different modes of death in the Cox regression analysis

HRV measurements that differed significantly between patients with any mode of death and the other patients in univariate comparison were tested one at a time in the clinical multivariate Cox hazard model (Table 2). Lower values of ln(LF) and DFA1 retained significant association with SCD/SCA after adjustments. However, ln(LF) was also associated with all other modes of death and DFA1 with NCD after adjustments. After relevant multivariate adjustments, SDNN and ln(VLF) remained associated with NSCD and NCD and LF/HF ratio with NCD. The PL slope lost its significant association with the occurrence of different modes of death after relevant adjustment in the clinical multivariate model. Cumulative proportional probabilities of SCD, NSCD and NCD in relation to SDNN, ln(LF) and DFA1 are shown in Fig 1. The cut-off points were obtained from receiver operating characteristics curves.

thumbnail
Fig 1. Cumulative proportional probabilities for sudden cardiac death or sudden cardiac arrest (SCD/SCA), non-sudden cardiac death (NSCD), and non-cardiac death (NCD).

https://doi.org/10.1371/journal.pone.0254107.g001

The cut-off points for the standard deviation of RR intervals (SDNN), the natural logarithm of low-frequency power of heart rate variability power spectrum (ln(LF)), and the short-term scaling exponent of the detrended fluctuation analysis (DFA1) were optimized from receiver operating characteristics curves. The log rank test was used to assess the statistical significance of the separations of the curves.

Association of heart rate variability measurements with clinical risk indicators

Associations of heart rate variability measurements with clinical risk indicators are shown in Table 3. Lower values of SDNN, ln(LF) and DFA1 were associated with higher age and lower LVEF, and lower values of DFA1 with larger LVMI. Females and patients with DM or CCS class ≥ 2 had lower values of SDNN, ln(LF) and DFA1, and smokers had lower values of SDNN.

thumbnail
Table 3. Association of heart rate variability measurements with clinical characteristics.

https://doi.org/10.1371/journal.pone.0254107.t003

Association of heart rate variability measurements with different modes of death in patients with diabetes

In the subgroup analysis of 729 diabetic patients, a total of 161 (22.1%) patients suffered death. Of the deaths, 41 (5.6%) were caused by SCD/SCA and 39 (5.3%) by NSCD. Death attributable to other causes occurred in 81 (11.1%) patients. Several clinical variables differed significantly in univariate comparisons between the study groups, and after adjustments, in the multivariate Cox regression analysis (Table 4). When left ventricular end-diastolic diameter, left ventricular end-systolic diameter, LVEF and LVMI were tested in the Cox multivariate clinical model in a stepwise manner, LVEF and LVMI retained a significant association with SCD/SCA. When age, alcohol consumption, CCS class ≥2 and LVMI were entered into the Cox multivariate clinical model, age and CCS class ≥ 2 retained a significant association with NSCD. When age, gender, body mass index, alcohol consumption, right bundle branch block and left ventricular end-diastolic diameter were tested in the Cox multivariate clinical model, age, gender and right bundle branch block retained a significant association with NCD. HRV measurements were entered one at a time into the multivariate clinical model as a continuous variable (Table 5). SDNN retained significant association with risk of SCD/SCA and NSCD, ln(LF) and ln(VLF) with all modes of death, and DFA1 only with SCD/SCA. LF/HF ratio and PL slope lost their significant association with NSCD and NCD after relevant adjustments.

thumbnail
Table 4. Clinical characteristics and heart rate variability parameters in study patients with diabetes at baseline.

https://doi.org/10.1371/journal.pone.0254107.t004

thumbnail
Table 5. Association of heart rate variability parameters with sudden cardiac death or sudden cardiac arrest, non-sudden cardiac death and non-cardiac death in the Cox regression analysis in the patients with diabetes.

https://doi.org/10.1371/journal.pone.0254107.t005

Discussion

The main finding of the present study was that several domains of HRV predicted an increased risk of SCD/SCA, NSCD and NCD in patients with CAD in the current treatment era. DFA1 and ln(LF) were associated with increased risk of SCD/SCA after adjustments with relevant clinical variables, but neither of the domains predicted a distinctively elevated risk of this mode of death. Generally, the HRV parameters failed to identify increased risk between a specific type of cardiac death and NCD, turning them into more overall predictors of death. Therefore, the clinical usefulness of HRV is limited in CAD patients in the current treatment era.

During the past decades, CHD mortality has decreased substantially almost all over the world, particularly in high-income countries [28]. Over these decades, breakthroughs have been made in evidence-based therapies, including new medicine and invasive treatment strategies, which have had a significant impact on decreasing mortality. It is estimated that approximately 45% of the decline in CHD mortality is attributable to medical and interventional/surgical treatments, of which secondary prevention and initial treatments for acute myocardial infarction or unstable angina have been among the most significant factors [19,20]. Nowadays, medications such as anti-thrombotic agents, lipid-lowering agents and beta-blockers are used extensively due to robust evidence of their favorable prognostic and anti-anginal influence [29,30]. Beta-blockers have an augmenting impact on HRV, which may be one of its action mechanisms and might decrease the predictive value of autonomic markers [31,32]. The vast majority of the patients in the present study were on beta-blocker medication. Percutaneous coronary intervention (PCI) and coronary artery bypass graft (CABG) are invasive treatment strategies which are routinely used in modern clinical settings and known to improve patients’ outcome and relive ischemic symptoms [29,30]. There is evidence that restored myocardial perfusion by PCI in CAD patients depresses HRV values instantly after operation, but HRV returns to the preoperative level and seems to trend beyond it [33,34]. After CABG, patients have permanently suppressed HRV levels but some domains of HRV return to preoperative values [35,36]. Therefore, it is suggested that HRV parameters may lose their predictive value after GABG due to HRV depressing and simultaneous survival improving effect. These influences are contradictory if HRV is used for risk stratification [37,38]. In the present study, a majority of the patients had prior PCI or CABG.

As discussed above, many of the modern evidence-based medical strategies have an effect on HRV parameters, whereas the risk of CHD mortality has been considerably reduced by using optimal strategies. Therefore, the association between HRV parameters and risk of SCD/SCA or NSCD in CAD patients might be relatively weaker in the current medical settings, which may partly explain why reduced HRV seems to be a more general indicator of mortality risk. Furthermore, it is plausible to speculate that patients with preponderant, particularly end stage, non-cardiac illness and increased cardiovascular risk profile represented by reduced HRV may die earlier [39].

Several previous studies have shown that reduced HRV predicts mortality in CAD patients with a prior myocardial infarction [5,10]. In a large study of post-myocardial infarction patients treated according to the modern guidelines, decreased HRV was found to predict both SCD and NSCD, an observation which is in line with our present findings in patients with stable CAD [40]. HRV measurements have also been observed to predict ventricular fibrillation or symptomatic sustained ventricular tachycardia in postinfarction patients [41]. Although there are some previous suggestions that nonlinear measurements of HRV, such as DFA1, could yield incremental prognostic information in postinfarction patients [12,13] and have a closer association with life-threatening ventricular tachyarrhythmias [15,42], none of the HRV measurements have been proven to be specific risk markers for SCD. In alignment with this concept, our present findings indicate that decreased HRV is rather a general risk marker for death than a specific indicator for the risk of SCD in the modern treatment era, and is therefore not helpful as a sole risk indicator for targeting ICD therapy to prevent SCD. Furthermore, patients with CAD need the best standard therapy including optimal medical therapy, and PCI or CABG, if needed, regardless of whether they have an increased risk profile or not as evaluated by HRV. Theoretically, it is possible that HRV could refine the risk for SCD if combined with information on other risk indicators including repolarization variability, such as the QT variability index [43]. In the present study, ln(LF) and DFA1 predicted SCD even after relevant adjustments; however, ln(LF) was also associated with risk of NSCD and NCD, and DFA1 with risk of NCD. Interestingly, concomitant activation of both vagal and sympathetic outflow typically decreases DFA1 [14], and some studies have suggested that an increase of sympathetic activity also can manifest as a decrease of the value of LF [44]. These autonomic nervous influences may increase the vulnerability to life-threatening ventricular tachyarrhythmias [45].

The association between reduced HRV and the risk for long-term mortality has been shown to be at least as strong in diabetic patients as in nondiabetic patients [46], an observation which is supported by our present findings. However, there has also been some controversy [38]. Decreased HRV values have been found in patients with diabetic neuropathy [47]. Concurring with these previous observations, the present patients with DM had lower values of HRV. It has earlier been shown that CAD patients with DM are at greater risk of SCD/SCA [21]. In the present analysis, decreased DFA1 values in patients with DM had closer specific association with the risk for SCD/SCA after relevant adjustments than the DFA1 values in all study patients.

Our study has some limitations. Our analysis was based on baseline HRV measurements. We did not perform control measurements of HRV during the follow-up. Therefore we were unable to evaluate, for example, aging-related changes in HRV, also considering that there were some differences in individual follow-up times.

Different domains of HRV are associated with increased risk of SCD, NSCD and NCD, but not distinctively with SCD in patients with CAD in the current treatment era. The clinical usefulness of HRV measurements alone in these patients is therefore limited.

Supporting information

References

  1. 1. Benjamin EJ, Muntner P, Alonso A, Bittencourt MS, Callaway CW, Carson AP, et al. Heart disease and stroke statistics-2019 update: a report from the American Heart Association. Circulation. 2019;139:e56–e528. pmid:30700139
  2. 2. Timmis A, Townsend N, Gale CP, Torbica A, Lettino M, Petersen SE, et al. European Society of Cardiology: cardiovascular disease statistics 2019. Eur Heart J. 2020;41:12–85. pmid:31820000
  3. 3. Myerburg RJ, Junttila MJ. Sudden cardiac death caused by coronary heart disease. Circulation. 2012;125:1043–1052. pmid:22371442
  4. 4. Carney RM, Saunders RD, Freedland KE, Stein P, Rich MW, Jaffe AS. Association of depression with reduced heart rate variability in coronary artery disease. Am J Cardiol. 1995;76:562–564. pmid:7677077
  5. 5. Kleiger RE, Miller JP, Bigger JT Jr, Moss AJ. Decreased heart rate variability and its association with increased mortality after acute myocardial infarction. Am J Cardiol. 1987;59:256–262. pmid:3812275
  6. 6. Tsuji H, Larson MG, Venditti FJ, Manders ES, Evans JC, Feldman CL, et al. Impact of reduced heart rate variability on risk for cardiac events. Circulation. 1996;94:2850–2855. pmid:8941112
  7. 7. Randall DC, Brown DR, Raisch RM, Yingling JD, Randall WC. SA nodal parasympathectomy delineates autonomic control of heart rate power spectrum. Am J Physiol Heart Circ Physiol. 1991;260:H985–H988. pmid:1672056
  8. 8. Vaishnav S, Stevenson R, Marchant B, Lagi K, Ranjadayalan K, Timmis AD. Relation between heart rate variability early after acute myocardial infarction and long-term mortality. Am J Cardiol. 1994;73:653–657. pmid:8166060
  9. 9. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Heart rate variability: standards of measurement, physiological interpretation and clinical use. Circulation. 1996;93:1043–1065. pmid:8598068
  10. 10. Bigger JT Jr, Fleiss JL, Steinman RC, Rolnitzky LM, Kleiger RE, Rottman JN. Frequency domain measures of heart period variability and mortality after myocardial infarction. Circulation. 1992;85:164–171. pmid:1728446
  11. 11. Bigger JT Jr, Steinman RC, Rolnitzky LM, Fleiss JL, Albrecht P, Cohen RJ. Power law behavior of RR-interval variability in healthy middle-aged persons, patients with recent acute myocardial infarction, and patients with heart transplants. Circulation. 1996;93:2142–2151. pmid:8925583
  12. 12. Mäkikallio TH, Høiber S, Køber L, Torp-Pedersen C, Peng CK, Goldberger AL, et al. Fractal analysis of heart rate dynamics as a predictor of mortality in patients with depressed left ventricular function after acute myocardial infarction. Am J Cardiol. 1999;83:836–839. pmid:10190395
  13. 13. Huikuri HV, Mäkikallio TH, Peng CK, Goldberger AL, Hintze U, Møller M. Fractal correlation properties of R-R interval dynamics and mortality in patients with depressed left ventricular function after an acute myocardial infarction. Circulation. 2000;101:47–53. pmid:10618303
  14. 14. Tulppo MP, Kiviniemi AM, Hautala AJ, Kallio M, Seppänen T, Mäkikallio TH, et al. Physiological background of the loss of fractal heart rate dynamics. Circulation. 2005;112:314–319. pmid:16009791
  15. 15. Mäkikallio TH, Koistinen J, Jordaens L, Tulppo MP, Wood N, Golosarsky B, et al. Heart rate dynamics before spontaneous onset of ventricular fibrillation in patients with healed myocardial infarcts. Am J Cardiol. 1999;83:880–884. pmid:10190403
  16. 16. Saul JP, Albrecht P, Berger RD, Cohen RJ. Analysis of long term heart rate variability: methods, 1/f scaling and implications. Comput Cardiol. 1987;14:419–422.
  17. 17. Vikman S, Mäkikallio TH, Yli-Mäyrä S, Pikkujämsä S, Koivisto AM, Reinikainen P, et al. Altered complexity and correlation properties of R-R interval dynamics before the spontaneous onset of paroxysmal atrial fibrillation. Circulation. 1999;100:2079–2084. pmid:10562264
  18. 18. Hogue CW Jr, Domitrovich PP, Stein PK, Despotis GD, Re L, Schuessler RB, et al. RR interval dynamics before atrial fibrillation in patients after coronary artery bypass graft surgery. Circulation. 1998;98:429–434. pmid:9714093
  19. 19. Unal B, Critchley JA, Capewell S. Explaining the decline in coronary heart disease mortality in England and Wales between 1981 and 2000. Circulation. 2004;109:1101–1107. pmid:14993137
  20. 20. Ford ES, Ajani UA, Croft JB, Critchley JA, Labarthe DR, Kottke TE, et al. Explaining the decrease in U.S. deaths from coronary disease, 1980–2000. N Engl J Med. 2007;356:2388–2398. pmid:17554120
  21. 21. Junttila MJ, Kiviniemi AM, Lepojärvi ES, Tulppo M, Piira OP, Kenttä T, et al. Type 2 diabetes and coronary artery disease: preserved ejection fraction and sudden cardiac death. Heart Rhythm. 2018;15:1450–1456. pmid:30274618
  22. 22. Kiviniemi AM, Lepojärvi ES, Tulppo MP, Piira OP, Kenttä TV, Perkiömäki JS, et al. Prediabetes and risk of cardiac death among patients with coronary artery disease: the ARTEMIS study. Diabetes Care. 2019;42:1319–1325. pmid:31076416
  23. 23. World Health Organization. Definition and diagnosis of diabetes mellitus and intermediate hyperglycemia: report of a WHO/IDF consultation. Genave: WHO Document Production Services; 2006.
  24. 24. Salo MA, Huikuri HV, Seppänen T. Ectopic beats in heart rate variability analysis: effects of editing on time and frequency domain measures. Ann Noninvasive Electrocardiol. 2001;6:5–17. pmid:11174857
  25. 25. Peng CK, Havlin S, Stanley HE, Goldberger AL. Quantification of scaling exponents and crossover phenomena in nonstationary heartbeat time series. Chaos. 1995;5:82–87. pmid:11538314
  26. 26. Pincus SM. Approximate entropy as a measure of system complexity. Proc Natl Acad Sci U S A. 1991;88:2297–2301. pmid:11607165
  27. 27. Lang RM, Badano LP, Mor-Avi V, Afilalo J, Armstrong A, Ernande L, et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. Eur Heart J Cardiovasc Imaging. 2015;16:233–271. pmid:25712077
  28. 28. Moran AE, Forouzanfar MH, Roth GA, Mensah GA, Ezzati M, Murray CJL, et al. Temporal trends in ischemic heart disease mortality in 21 world regions, 1980 to 2010. Circulation. 2014;129:1483–1492. pmid:24573352
  29. 29. Fihn SD, Gardin JM, Abrams J, Berra K, Blankenship JC, Dallas AP, et al. 2012 ACCF/AHA/ACP/AATS/PCNA/SCAI/STS guideline for the diagnosis and management of patients with stable ischemic heart disease. J Am Coll Cardiol. 2012;60: e44–e164. pmid:23182125
  30. 30. Knuuti J, Wijns W, Saraste A, Capodanno D, Barbato E, Funck-Brentano C, et al. 2019 ESC guidelines for the diagnosis and management of chronic coronary artery syndromes. Eur Heart J. 2020;41:407–477. pmid:31504439
  31. 31. Niemelä MJ, Airaksinen KEJ, Huikuri HV. Effect of beta-blockade on heart rate variability in patients with coronary artery disease. J Am Coll Cardiol. 1994;23:1370–1377. pmid:8176095
  32. 32. Lampert R, Ickovics JR, Viscoli CJ, Horwitz RI, Lee FA. Effects of propranolol on recovery of heart rate variability following acute myocardial infarction and relation to outcome in the Beta-Blocker Heart Attack Trial. Am J Cardiol. 2003;91:137–142. pmid:12521623
  33. 33. Wennerblom B, Lurje L, Solem J, Tygesen H, Udén M, Vahisalo R, et al. Reduced heart rate variability in ischemic heart disease is only partially caused by ischemia. Cardiology. 2000;94:146–151. pmid:11279319
  34. 34. Kanadasi M, Kudaiberdieva G, Birand A. Effect of the final coronary arterial diameter after coronary angioplasty on heart rate variability responses. Ann Noninvasive Electrocardiol. 2002;7:106–113. pmid:12049681
  35. 35. Cygankiewicz I, Wranicz JK, Bolinska H, Zaslonka J, Jaszewski R, Zareba W. Influence of coronary artery bypass grafting on heart rate turbulence parameters. Am J Cardiol. 2004;94:186–189. pmid:15246897
  36. 36. Laitio TT, Huikuri HV, Koskenvuo J, Jalonen J, Mäkikallio TH, Helenius H, et al. Long-term alterations of heart rate dynamics after coronary artery bypass graft surgery. Anesth Analg. 2006;102:1026–1031. pmid:16551892
  37. 37. Milicevic G, Fort L, Majsec M, Bakula V. Heart rate variability decreased by coronary artery surgery has no prognostic value. Eur J Cardiovasc Prev Rehabil. 2004;11:228–232. pmid:15179105
  38. 38. Stein PK, Domitrovich PP, Kleiger RE. Including patients with diabetes mellitus or coronary artery bypass grafting decreases the association between heart rate variability and mortality after myocardial infarction. Am Heart J. 2004;147:309–316. pmid:14760330
  39. 39. Perkiömäki JS, Ruwald AC, Kutyifa V, Ruwald MH, Mcnitt S, Polonsky B, et al. Risk factors and the effect of cardiac resynchronization therapy on cardiac and non-cardiac mortality in MADIT-CRT. Europace. 2015;17:1816–1822. pmid:26071234
  40. 40. Mäkikallio TH, Barthel P, Schneider R, Bauer A, Tapanainen JM, Tulppo MP, et al. Prediction of sudden cardiac death after acute myocardial infarction: role of Holter monitoring in the modern treatment era. Eur Heart J. 2005;26:762–769. pmid:15778204
  41. 41. Huikuri HV, Raatikainen MJP, Moerch-Joergensen R, Hartikainen J, Virtanen V, Boland J, et al. Prediction of fatal or near-fatal cardiac arrhythmia events in patients with depressed left ventricular function after an acute myocardial infarction. Eur Heart J. 2009;30:689–698. pmid:19155249
  42. 42. Mäkikallio TH, Seppänen T, Airaksinen KE, Koistinen J, Tulppo MP, Peng CK, et al. Dynamic analysis of heart rate may predict subsequent ventricular tachycardia after myocardial infarction. Am J Cardiol. 1997;80:779–783. pmid:9315590
  43. 43. Haigney MC, Zareba W, Gentlesk PJ, Goldstein RE, Illovsky M, McNitt S, et al. QT interval variability and spontaneous ventricular tachycardia or fibrillation in the multicenter automatic defibrillation implantation trial (MADIT) II patients. J Am Coll Cardiol. 2004;44:1481–1487. pmid:15464332
  44. 44. Koivikko ML, Tulppo MP, Kiviniemi AM, Kallio MA, Perkiömäki JS, Salmela PI, et al. Autonomic cardiac regulation during spontaneous nocturnal hypoglycemia in patients with type 1 diabetes. Diabetes Care. 2012;35:1585–1590. pmid:22611064
  45. 45. Shusterman V, Aysin B, Gottipaty V, Weiss R, Brode S, Schwartzman D, et al. Autonomic nervous system activity and the spontaneous initiation of ventricular tachycardia. J Am Coll Cardiol. 1998;32:1891–1899. pmid:9857869
  46. 46. Whang W, Bigger JT Jr. Comparison of the prognostic value of RR-interval variability after acute myocardial infarction in patients with versus those without diabetes mellitus. Am J Cardiol. 2003;92:247–251. pmid:12888125
  47. 47. Pfeifer MA, Cook D, Brodsky J, Tice D, Reenan A, Swedine S, et al. Quantitative evaluation of cardiac parasympathetic activity in normal and diabetic man. Diabetes. 1982;31:339–345. pmid:7152130