- Email: [email protected]
Cardiorespiratory Fitness and Risk of Sudden Cardiac Death in Men and Women in the United States
ORIGINAL ARTICLE
Cardiorespiratory Fitness and Risk of Sudden Cardiac Death in Men and Women in the United States: A Prospective Evaluation From the Aerobics Center Longitudinal Study David Jiménez-Pavón, PhD; Enrique G. Artero, PhD; Duck-chul Lee, PhD; Vanesa España-Romero, PhD; Xuemei Sui, MD, PhD; Russell R. Pate, PhD; Timothy S. Church, MD, PhD; Luis A. Moreno, MD, PhD; Carl J. Lavie, MD; and Steven N. Blair, PED Abstract Objectives: To examine the relation between cardiorespiratory fitness (CRF) and sudden cardiac death (SCD) in a large US adult population and to study the effects of hypertension, obesity, and health status on the relation of CRF with SCD. Patients and Methods: A total of 55,456 individuals (mean age, 44.2 years; 13,507 women) from the Aerobics Center Longitudinal Study, a prospective observational investigation (from January 2, 1974, through December 31, 2002), were included. Cardiorespiratory fitness was assessed by a maximal treadmill test, and baseline assessment included an extensive set of measurements. Results: There were 109 SCDs. An inverse risk of SCD was found across incremental CRF levels after adjusting for potential confounders. Participants with moderate and high CRF levels had 44% (hazard ratio, 0.56; 95% CI, 0.35-0.90) and 48% (hazard ratio, 0.52; 95% CI, 0.30-0.92) significantly lower risk of SCD, respectively, than did those with low CRF levels (P<.001). The risk of SCD decreased by 14% (hazard ratio, 0.86; 95% CI, 0.77-0.96) per 1-metabolic equivalent increase in the fully adjusted model. Hypertensive, overweight, or unhealthy individuals with moderate to high CRF levels had lower risks of SCD (ranging from 58% to 72% of lower risk) than did those with the same medical conditions and low CRF levels. Conclusion: The risk of SCD in US men and women could be partially reduced by ensuring moderate to high levels of CRF independently of other risk factors and especially in those who are hypertensive, overweight, or unhealthy. ª 2016 Mayo Foundation for Medical Education and Research
S
udden cardiac death (SCD) is recognized as a relevant cause of mortality, which in many cases occurs in the general population without previous symptoms of any coronary heart disease (CHD).1-3 Although 70% to 90% of the total incidence of SCD occurs in men, this cause of mortality has been documented in both sexes in individuals with a history of CHD or other major cardiovascular (CV) disease (CVD) risk factors and in those without a history of CHD or CVD.2,4 Therefore, SCD constitutes an important public health problem with multiple risk factors.5 Although numerous studies6-8 have attempted to identify those factors that are
n
Mayo Clin Proc. 2016;91(7):849-857
associated with higher SCD risk for the general population, conclusive evidence on this point remains somewhat elusive. A moderate to high level of cardiorespiratory fitness (CRF) has been associated with many CV benefits and has been established as a predictor of all-cause mortality in men and women.9-12 Moreover, the use of treadmill protocols and estimation of workloads based on treadmill speeds and inclines, typically reported as estimated metabolic equivalents (METs), as an indirect measurement of oxygen consumption, has widely reported to accurately predict these associations.9-12 However, only 1 study13 focused on the relation between
Mayo Clin Proc. n July 2016;91(7):849-857 n http://dx.doi.org/10.1016/j.mayocp.2016.04.025 www.mayoclinicproceedings.org n ª 2016 Mayo Foundation for Medical Education and Research
From the GALENO Research Group, Department of Physical Education, University of Cádiz, Puerto Real, Cádiz, Spain (D.J.-P., V.E.-R.); Department of Exercise Science (D.J.-P., E.G.A., X.S., R.R.P., S.N.B.) and Department of Epidemiology and Biostatistics (S.N.B.), University of South Carolina, Columbia; GENUD (Growth, Exercise, Affiliations continued at the end of this article.
849
MAYO CLINIC PROCEEDINGS
CRF and SCD, in which authors studied 2368 European middle-aged men and reported that CRF (cycle ergometer test) was inversely related to the risk of SCD. In addition, other authors studied the effect of specific risk factors, such as systemic blood pressure (BP) and hypertension (HTN)14 and asymptomatic ST-segment depression15 in men or a cluster of lifestyle factors in women,16 on the risk of SCD. One of the American Heart Association’s 2020 Impact Goals17 is to decrease CVD mortality. Thus, prevention of SCD by detecting early risk factors could be a useful tool contributing to the American Heart Association’s goal through decreasing the incidence of SCD at the population level. The identification of these risk factors would facilitate the large-scale screening of those participants at higher risk of SCD. However, to our knowledge, there has been no large-scale prospective assessment of the relation between CRF and SCD, including men and women, and assessment of the effect of CRF, especially combined with the other important CVD risk factors. Therefore, the present study aimed to examine the relation between CRF and SCD in a large population of men and women in the United States. We also sought to study the specific effects of HTN, obesity, and overall health status on the relation of CRF with SCD. PATIENTS AND METHODS The present report is based on data from the Aerobics Center Longitudinal Study, a prospective observational investigation into the association of clinical and lifestyle factors with disease and health outcomes in patients examined at the Cooper Clinic in Dallas, Texas.11,17 Participants came to the clinic for periodic preventive health examinations and for counseling on diet, exercise, and other lifestyle factors associated with an increased risk of chronic disease. Participants were volunteers, not paid, and were not recruited into the study, but sent by their employers or physicians or were self-referred for the examination. The Cooper Institute’s institutional review board reviewed and approved the study protocol annually. For the present analysis, participants 20 years and older with complete data on CRF and a group of potential confounders including 850
Mayo Clin Proc.
n
health behaviors, blood parameters, and disease-related information were included. From a total of 59,611 participants, we excluded those who failed to achieve at least the 85% of age-predicted maximum heart rate during the treadmill test (n¼1868) or were followed for less than 1 year (n¼2287). These criteria resulted in 55,456 individuals (13,507 women) with a mean age of 44.2 years (ranging from 20 to 100 years), whose baseline examination occurred from January 2, 1974, through December 31, 2003. Participants were predominantly white, well-educated, and belonged to the middle and upper socioeconomic strata. Outcomes and Follow-Up Procedures of the clinical examinations are described in detail elsewhere.11,18 Briefly, the baseline assessment was performed after an overnight fast and included an extensive physical examination and an array of clinical measurements. Body mass index (BMI) was calculated as the weight in kilograms divided by the height in meters squared (kg/m2) and measured using a standard clinical scale and stadiometer. Overweight and obesity were defined as a BMI of 25 to 29.9 and 30 or more kg/m2, respectively. Resting systolic and diastolic BP was measured in the seated position as the first and fifth Korotkoff sounds by using standard auscultation methods after at least 5 minutes of sitting quietly and recorded as the average of at least 2 readings separately by 2 minutes.19 If the first 2 readings differed by more than 5 mm Hg, additional readings were obtained and averaged; HTN was defined as systolic or diastolic BP of 140/90 mm Hg or greater or a history of physician diagnosed HTN. Concentrations of total cholesterol and fasting glucose were measured using automated techniques in accordance with the standards of the Centers for Disease Control and Prevention lipid standardization program. Hypercholesterolemia was defined as a serum total cholesterol level of 240 mg/dL or greater (to convert to mmol/L, multiply by 0.0259). Diabetes mellitus was defined according to the American Diabetes Association criteria,20 that is, a fasting plasma glucose level of 126 mg/dL or greater, use of insulin, or self-report of previous physician diagnosis. Participants completed a standardized questionnaire on medical history that
July 2016;91(7):849-857
n
http://dx.doi.org/10.1016/j.mayocp.2016.04.025 www.mayoclinicproceedings.org
CARDIORESPIRATORY FITNESS AND SUDDEN CARDIAC DEATH
included a personal history of myocardial infarction, stroke, HTN, diabetes mellitus, and cancer; a family history of CVD; smoking status (never, former, or current smoker); alcohol intake (heavy drinker or not, defined as >14 and 7 drinks/wk for men and women, respectively; 1 unit of alcohol was defined as 12 oz of beer, 5 oz of wine, or 1.5 oz of hard liquor); and physical activity (active or inactive). Physical inactivity was defined as reporting no physical activity during leisure time in the 3 months before the baseline examination. Electrocardiography showing abnormalities (both at rest and exercise) were largely defined as rhythm and conduction disturbances and ischemic ST-T wave abnormalities, as described elsewhere.21 Cardiorespiratory Fitness. Cardiorespiratory fitness was assessed by a maximum treadmill exercise test using a modified Balke protocol.11,22 Participants began walking at a speed of 88 m/min without elevation. After the first minute, the elevation was increased by 2%; thereafter, the elevation was increased by 1% per minute until the 25th minute. After 25 minutes, the elevation did not change but the speed was increased by 5.4 m/min each minute until the test end point. Participants were encouraged to give maximal effort, and the test was stopped when they reached volition exhaustion or the doctor intervened for medical reasons. The mean ageepredicted maximum heart rate (220age) achieved during exercise was used to determine individuals who achieved maximal effort. Exercise duration using this protocol is highly correlated with measured maximum oxygen uptake (r0.90) in both men23 and women.24 To standardize the interpretation of exercise test performance, maximal METs (1 MET¼3.5 mL oxygen uptake/kg per minute) were estimated on the basis of the final treadmill speed and grade.25 Cardiorespiratory fitness was categorized as low, moderate, or high age- and sex-specific distributions of maximal exercise duration from the entire Aerobics Center Longitudinal Study population according to previous publications.11,18 Sudden Cardiac Death. All participants were followed from the date of their baseline examination until their date of death or until Mayo Clin Proc. n July 2016;91(7):849-857 www.mayoclinicproceedings.org
n
December 31, 2003. We computed personyears of exposure as the sum of follow-up time in decedents and survivors. The National Death Index was the primary data source for mortality surveillance, augmented with death certificates as previously used.13 The National Death Index is an accurate method of ascertaining deaths in observational studies, with high sensitivity (96%) and specificity (100%).25 The underlying and contributing cause of death was determined from the index or by a nosologist’s review of official death certificates obtained from the department of vital records in the decedent’s state of residence. Sudden cardiac death was defined by the International Classification of Diseases, Ninth Revision code 798.1 before 1999 and by the International Classification of Diseases, Tenth Revision code I46 between 1999 and 2003 according to others.13 Statistical Analyses Descriptive analyses summarized baseline characteristics of the participants according to CRF categories. Tests for linear trends across CRF categories were performed using general linear models for continuous variables and the chi-square test for categorical variables. Cox proportional hazards regression was used to estimate hazard ratios (HRs), 95% CIs, and mortality rates (deaths per 10,000 person-years of follow-up) according to CRF categories. No significant effect modification by sex was observed on SCD using interaction terms (CRFsex) in Cox regressions; thus, the analyses were performed on the pooled sample. Multivariable analyses according to 3 models of adjustments were performed: model 1 was the unadjusted model; model 2 included adjustments for age and sex; and model 3 included adjustments for model 2 plus examination year, BMI, current smoking, heavy alcohol intake (>14 or 7 drinks/wk for men and women, respectively), diabetes mellitus (presence or absence), HTN (presence or absence), electrocardiogram showing abnormalities at rest or exercise (presence or absence), and parental history of CVD. Further analyses were performed to estimate the hazards for SCD by aerobic capacity using METs as a continuous variable and using the same models of adjustments.
http://dx.doi.org/10.1016/j.mayocp.2016.04.025
851
MAYO CLINIC PROCEEDINGS
In addition, the effects of HTN, obesity, and health status (all presence or absence) on the association of CRF with risk of SCD were explored. Health status was defined as the presence or absence of electrocardiogram showing abnormalities and personal history of CVD or cancer. The analyses were adjusted for age, sex, examination year, and electrocardiogram showing abnormalities (only for the presence or absence of HTN) according to the number of SCDs per group. Further analyses were performed to estimate the effect of health status on the hazards for SCD by METs.
RESULTS Descriptive Characteristics There were 109 SCDs (20 per 10,000 personyears) in men and women during an average follow-up of 14.7 years. Table 1 summarizes the baseline characteristics of the participants according to CRF categories. Body mass index, total cholesterol level, fasting blood glucose level, and BP were lower in those with higher CRF levels. The prevalence of participants who were inactive, current smokers, and having hypercholesterolemia, diabetes mellitus, HTN, electrocardiogram showing abnormalities,
TABLE 1. Baseline Characteristics of the Participants from the Aerobics Center Longitudinal Study (1974-2003) According to the Cardiorespiratory Fitness Levels and Sudden Cardiac Deatha,b Cardiorespiratory fitness level Characteristic
All (N¼55,456)
Low (n¼8404)
Moderate (n¼21,458)
High (n¼25,594)
P value for linear trendc
Age (y) Body mass index (kg/m2) Treadmill time (min) Maximum METs Total cholesterol level (mg/dL)d Fasting blood glucose level (mg/dL) Waist circumference (cm) Triglyceride level (mg/dL) Blood pressure (mm Hg) Systolic Diastolic Physically inactivee Current smokers Heavy drinkersf Hypercholesterolemiag Diabetes mellitush HTNi Electrocardiogram showing abnormalitiesj Personal history of CVD Cancerk Parental history of CVD Metabolic syndromel
44.210 25.74.1 16.85.2 11.12.5 206.239.7 98.816.8 87.219.6 128.4104.7
43.59.4 29.05.6 10.52.9 8.21.4 216.041.9 103.825.9 98.721.6 175.8141.9
44.39.9 26.23.7 14.93.1 10.21.4 209.839.8 99.216.3 90.118.8 139.3102.4
44.210.2 24.23.0 20.54.3 12.92.2 199.937.8 96.712.6 82.218.0 103.683.0
<.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001
119.714.4 80.09.9 16,019 (28.9) 8946 (16.1) 9801 (17.7) 14,938 (26.9) 2982 (5.4) 15,807 (28.5) 4237 (7.6) 658 (1.2) 1805 (3.3) 14,862 (26.8) 8964 (16.2)
123.314.9 83.010.3 5243 (62.4) 2401 (28.6) 1472 (17.5) 2853 (33.9) 834 (9.9) 3503 (41.7) 874 (10.4) 192 (2.3) 156 (1.9) 2461 (29.3) 2512 (29.9)
120.014.2 80.79.9 7915 (36.9) 4145 (19.3) 3771 (17.6) 6329 (29.5) 1225 (5.7) 6620 (30.9) 1719 (8.0) 282 (1.3) 626 (2.9) 5831 (27.2) 4354 (20.3)
118.314.1 78.69.4 2861 (11.2) 2400 (9.4) 4558 (17.8) 5756 (22.5) 923 (3.6) 5684 (22.2) 1644 (6.4) 184 (0.7) 1023 (4.0) 6570 (25.7) 2098 (8.2)
<.001 <.001 <.001 <.001 .74 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001
CVD ¼ cardiovascular disease; HTN ¼ hypertension. Data are presented as mean SD or as No. (percentage). c Differences across cardiorespiratory fitness categories using general linear models for continuous variables or the c2 test for categorical variables. d SI conversion factor: To convert mg/dL values to mmol/L, multiply by 0.0259. e Defined as reporting no physical activity during leisure time in the 3 mo before the examination. f Defined as >14 and 7 drinks/wk for men and women, respectively (1 unit of alcohol defined as 12 oz of beer, 5 oz of wine, and 1.5 oz of hard liquor). g Defined as total cholesterol level of 240 mg/dL or previous physician diagnosis. h Defined as fasting plasma glucose level of 126 mg/dL, previous physician diagnosis, or use of insulin. i Defined as resting blood pressure of 140/90 mm Hg or previous physician diagnosis. j Defined as rhythm and conduction disturbances and ischemic ST-T wave abnormalities (at rest or exercise). k Defined as personal history of cancer. l Metabolic syndrome established using the National Cholesterol Education Program Adult Treatment Panel III criteria and was based on the presence of 3 risk factors.26 a
b
852
Mayo Clin Proc.
n
July 2016;91(7):849-857
n
http://dx.doi.org/10.1016/j.mayocp.2016.04.025 www.mayoclinicproceedings.org
CARDIORESPIRATORY FITNESS AND SUDDEN CARDIAC DEATH
and parental history of CVD was lower at higher CRF levels. When corrections for multiple comparisons were applied, the results with P values less than .002 remained significant. Hazard Ratios for SCD by CRF Levels Table 2 lists the mortality rate and HR for SCD calculated using 3 models of adjustments. Mortality rates for SCD were lower at higher CRF levels. An inverse risk of SCD was found across incremental CRF levels in all the models (P value for linear trend: P<.05 for model 1 and P<.001 for models 2 and 3). Participants with moderate and high CRF levels had 44% (HR, 0.56; 95% CI, 0.35-0.90) and 48% (HR, 0.52; 95% CI, 0.30-0.92) significantly lower risk of SCD, respectively, than did those with low CRF levels in the multivariate-adjusted model (model 3). The risk of SCD decreased by 14% (HR, 0.86; 95% CI, 0.77-0.96) per 1-MET increase in the fully adjusted model (model 3). Effects of HTN, Obesity, and Health Status The Figure shows the risk of SCD across CRF levels by the presence or absence of HTN (panel A), by the presence or absence of normal weight or overweight or obese (panel B), and in healthy or unhealthy groups (panel C) after adjustments for age, sex, examination year, and electrocardiogram showing abnormalities (only for the presence or absence of HTN). Inverse trends in the risk of SCD across CRF levels were observed in all subgroups, but only reaching statistical significance in HTN (and the normotensive group with high CRF
levels), overweight or obese, and healthy or unhealthy (moderate CRF levels) groups. Those participants identified with HTN and with moderate or high CRF levels had 65% (HR, 0.35; 95% CI, 0.18e0.67) and 72% (HR, 0.28; 95% CI, 0.12e0.61) lower risk of SCD, respectively, than did those with low CRF levels. The normotensive participants with high CRF levels had 51% (HR, 0.49; 95% CI, 0.24e0.98) lower risk of SCD than did those with low CRF levels. Similarly, the risk of SCD was lower in overweight or obese participants with moderate or high CRF levels (HR, 0.42; 95% CI, 0.25e0.72 and HR, 0.36; 95% CI, 0.18e0.71, respectively). The healthy group had a mean CRF level of 11 METs vs the unhealthy group with 10 METs (P<.001). In the unhealthy group, those participants with moderate CRF levels had 69% (HR, 0.31; 95% CI, 0.13e0.72) lower risk of SCD, whereas participants classified in the healthy group and with moderate or high CRF levels had 47% (HR, 0.53; 95% CI, 0.31e0.90) and 65% (HR, 0.35; 95% CI, 0.19e0.65), respectively, lower risk of SCD than did those with low CRF levels. Each 1-MET increase was related to 20% (HR, 0.80; 95% CI, 0.71e0.90; P¼.001) lower risk in the healthy group and to 18% (HR, 0.82; 95% CI, 0.68e0.98; P¼.05) lower risk in the unhealthy group. Sensitivity Analyses The results were similar when all the analyses were repeated using an older age group (age, 40 years; n¼36,720); specifically, CRF
TABLE 2. Rates and Hazards Ratios for SCD in Men and Women According to Cardiorespiratory Fitness Levelsa
Variable Cardiorespiratory fitness level Low Moderate High P value for linear trend Per 1-MET increase
No. of SCDs (n)
Mortality rate (%)b
38 (8366) 41 (21,417) 30 (25,564)
2.6 1.2 0.9
Hazard ratio (95% CI) Model 1
Model 2
Model 3
1.00 (reference) 0.62 (0.40-0.96) 0.56 (0.35-0.91) .03 0.77 (0.70-0.84)
1.00 (reference) 0.52 (0.33-0.80) 0.40 (0.25-0.66) <.001 0.82 (0.75-0.90)
1.00 (reference) 0.56 (0.35-0.90) 0.52 (0.30-0.92) <.001 0.86 (0.77-0.96)
MET ¼ metabolic equivalent; n ¼ number of participants; SCD ¼ sudden cardiac death. Per 10,000 person-years, adjusted for age, sex, and examination year. Model 1 was unadjusted; model 2 was adjusted for age and sex; model 3 was adjusted for model 2 plus examination year, body mass index, current smoking, heavy alcohol intake (>14 and 7 drinks/wk for men and women, respectively), diabetes mellitus (presence or absence), hypertension (presence or absence), electrocardiogram showing abnormalities (presence or absence), and parental history of cardiovascular disease.
a
b
Mayo Clin Proc. n July 2016;91(7):849-857 www.mayoclinicproceedings.org
n
http://dx.doi.org/10.1016/j.mayocp.2016.04.025
853
MAYO CLINIC PROCEEDINGS
mortality rates were 3.9%, 1.9%, and 1.3% for low, middle, and high CRF levels, respectively. Alternatively, Cox regression models using personal history of CVD instead of electrocardiogram showing abnormalities did not modify the main results. Furthermore, including those who did not reach 85% of the age-predicted maximum heart rate did not significantly change the main findings. Moreover, the incidence of SCD (reported as a percentage of allcause mortality) across fitness levels and main risk factors was similar to the incidence of SCD in the overall population (data not shown). Finally, the results were not materially altered when analyses were repeated excluding participants without SCD and including those with other causes of mortality.
Hazard ratio of sudden cardiac death
2.5
2.0
1.5 Reference
Reference
1.0 0.62
0.5
0.49 0.35
0.28
0.0 Low
Moderate High CRF No. of deaths 15 26 21 (No. of participants) (4901) (14,839) (19,910)
A
Low
Moderate High CRF 15 9 23 (3503) (6621) (5684)
Normotensive
Hypertensive
Hazard ratio of sudden cardiac death
2.5
2.0
1.5 Reference
Reference
1.0 0.57
0.5
0.42 0.42
0.36
0.0 Low
Moderate High CRF No. of deaths 7 15 17 (No. of participants) (1973) (8075) (15,932)
B
Normal weight
Low
Moderate High CRF 31 26 13 (6431) (13,385) (9662) Overweight and obese
Hazard ratio of sudden cardiac death
2.5
2.0
1.5 Reference
Reference
1.0
0.53
0.5
0.35
0.31
0.41
0.0 Low
Moderate High CRF 25 32 21 No. of deaths (No. of participants) (7223) (19,046) (22,940)
C
Healthy
Low
Moderate High CRF 13 9 9 (1081) (2414) (2654) Unhealthy
FIGURE. Association of cardiorespiratory fitness (CRF) (low, moderate, and high) with risk of sudden cardiac death (A) by the presence or absence of hypertension, (B) by the presence or absence of normal weight or overweight or obesity, and (C) in healthy or unhealthy groups, in which unhealthy was defined as participant having electrocardiogram showing abnormalities or personal history of cardiovascular disease or cancer. The analyses were adjusted for age, sex, examination year, and electrocardiogram showing abnormalities (only for the presence or absence of hypertension) according to the number of sudden cardiac deaths per group.
854
Mayo Clin Proc.
n
DISCUSSION The main finding from these prospective analyses was that participants with moderate to high CRF levels had about half the risk of SCD than did individuals with low CRF levels, and this was independent of other potentially confounding factors. Therefore, low CRF levels is a major risk factor for SCD in a large and general population of men and women in the United States This study also found that the decreases in the risk of SCD by having moderate to high CRF levels were even higher (ranging from 58% to 72% of lower risk) in the HTN, overweight or obese, and unhealthy participants. Each 1-MET increase in the level of CRF corresponded to an adjusted risk reduction in both healthy and unhealthy populations (20% and 18%, respectively). The only previous study addressing the relation between CRF and SCD was performed by Laukkanen et al,13 in which 2368 Finish men (age, 42-60 years) were studied during 17 years of follow-up by using a cycle ergometer test for the estimation of CRF (in METs). These authors13 found that each 1-MET increase in CRF levels was associated with a reduction in the risk of SCD of 22% in an adjusted risk model. Our study adds to this previous research by analyzing a larger population including both men and women from the United States and using a maximal treadmill exercise test for the estimation of CRF, which is the most used means of assessing exercise capacity in the United States and
July 2016;91(7):849-857
n
http://dx.doi.org/10.1016/j.mayocp.2016.04.025 www.mayoclinicproceedings.org
CARDIORESPIRATORY FITNESS AND SUDDEN CARDIAC DEATH
also recreates the natural skill of walking/ running as a tool for population-based screenings. Our results confirmed the previous findings and extend them to a large sample of US men and women, highlighting the importance of having moderate to high levels of CRF (10.2 and 12.9 METs, respectively) in relation to SCD. Other authors14-16 have also assessed SCD in relation to other independent or clustered risk factors. They reported that men with HTN (systolic BP >145 mm Hg; n¼2666) had a 2.04-fold higher adjusted risk for SCD than did those with systolic BP less than 123 mm Hg.14 Moreover, Laukkanen et al15 studied the association between silent ST-segment depression during and after the maximal exercise test and the risk of SCD in men (n¼1769) and found that ST-segment depression during and after the exercise test was a strong predictor of the risk of SCD, with special relevance in smokers, individuals with hypercholesterolemia, and men with HTN. Recently, Chiuve et al16 performed a prospective study with 81,722 American women during 26 years of follow-up, analyzing the degree to which adherence to a healthy lifestyle (as a cluster of 4 self-reported components: smoking, physical activity, diet, and weight) affected the risk of SCD, and found that all 4 low-risk factors were significantly and independently related to a lower risk of SCD in women. In our present study, the particular effects of other risk factors with regard to the relation of CRF and SCD were also assessed. We confirmed the relevance of these risk factors as reported in previous studies,14,16 but also provided a new approach from the viewpoint of CRF levels. In fact, we observed that CRF levels were related to a lower risk of SCD in both healthy and unhealthy men and women, but these relations were stronger in those participants with HTN, who were overweight or obese, and who were unhealthy. These results confirmed the previous findings on the relation of HTN, BMI, and healthy status with SCD,14,16 but also suggested that higher CRF levels may add additional strength to the associations with SCD in men and women with, but also without, CVD symptoms or risk factors. The understanding of the possible mechanism underlying the association of CRF with risk of SCD is not clear, and it cannot be Mayo Clin Proc. n July 2016;91(7):849-857 www.mayoclinicproceedings.org
n
confirmed in the present study. However, we hypothesize that similar reasons for the beneficial association of CRF with CHD-related mortality would also provide protection from SCD. In this regard, having a moderate or high level of CRF has been repeatedly shown to be associated with a lower risk of CVD morbidity and mortality in apparently healthy individuals, as well as in those with other CVD risk factors.9,11,12,21,27 Moreover, an important report by Jouven et al28 highlighted that alterations in the neural control of cardiac function contribute to the risk of SCD, and they reported that abnormal heart rate profiles during exercise and recovery were related to a higher risk of SCD.29-31 The CRF level has been found to affect the vagal modulation of heart rate as well as to positively affect the CV autonomic function. Consequently, having a moderate to high level of CRF could decrease the risk of SCD through the different benefits on the CV and nervous systems attributed to exercise and overall physical fitness. There are several limitations of our study that should be considered. The Aerobics Center Longitudinal Study study sample was mainly composed of white men and women; thus, as black populations have higher mortality rates for SCD than do whites or others,32 prospective studies on populations from different ethnicities are needed. Another limitation is that a detailed history on medication usage was not available. However, because the present study does not have a randomized controlled trial design, we cannot establish the dose-response effect of CRF increments on SCD. No information on the causes of SCD was reported, such as congenital heart disease or electrical or structural abnormalities of the heart and great vessels. Moreover, murmur was not included in the database, nor were echocardiograms routinely performed. However, because of the long follow-up necessary for this kind of study and the large scale of the sample size required, it would be extremely difficult to corroborate this hypothesis or perform more detailed assessments in a randomized study. Because previous publications reported associations of CRF with other mortality outcomes using this data set, but SCD was not explored as an outcome, we think these findings might be useful to understand the relation between
http://dx.doi.org/10.1016/j.mayocp.2016.04.025
855
MAYO CLINIC PROCEEDINGS
CRF and SCD. Finally, although our sample size (>55,000 and w25% women) and follow-up period (average, 14.7 years) were quite robust, the incidence of SCD was rare, and our analysis of only 109 SCD events somewhat limits the overall statistical power of our study. This relatively low event rate of SCD observed could be partly explained by the outcome classification. A death is classified as SCD when it occurred within 24 hours of the onset of symptoms (including nonwitnessed SCD cases as well) and there are not any other noncardiac causes of sudden death. If a death was witnessed, an unexpected cardiac event that occurs within 1 hour from the onset of symptoms is typically classified as SCD. In this sense, the use of International Classification of Diseases codes to identify SCD could have underestimated the overall number of deaths and all SCD cases may not have been included; however, the National Death Index has shown to be an accurate method of ascertaining deaths in observational studies25 and was previously used.13 Finally, the indirect use of CRF, as there was no exchange of gases during the test, could be another limitation; however, this indirect use of CRF has shown to be sensitive in previous studies.11,18 The present study also has several additional strengths, including the use of highly reliable measurement through a maximal treadmill exercise protocol to determine CRF levels, which allowed a better transference to habitual activities (walking or running) in comparison with cycling13 and to avoid potential bias derived from the use of a questionnaire.16 CONCLUSION The risk of SCD in US men and women is significantly lower in those having moderate to high levels of CRF independently of other risk factors and even in those who are asymptomatic. To highlight the importance of having a higher CRF level is the fact that in particularly vulnerable groups, such as hypertensive, obese, or unhealthy populations, the reduction in the risk of SCD when having moderate to high CRF levels could achieve the approximately 75% of the total risk. Therefore, not only is obtaining the CRF level a useful tool in screening for the risk of CVD and CVD mortality, it also provides additional information on the 856
Mayo Clin Proc.
n
risk of SCD in asymptomatic men and women. Furthermore, our data provide information on the potential benefits of moderate to high CRF levels, especially in higher risk patients, for the primary prevention of SCD.
ACKNOWLEDGMENTS We thank physicians and technicians at the Cooper Clinic for collecting the data and staff at the Cooper Institute for data entry and data management. Abbreviations and Acronyms: BMI = body mass index; BP = blood pressure; CHD = coronary heart disease; CRF = cardiorespiratory fitness; CV = cardiovascular; CVD = cardiovascular disease; HR = hazard ratio; HTN = hypertension; MET = metabolic equivalent; SCD = sudden cardiac death Affiliations (Continued from the first page of this article.): Nutrition and Development) Research Group, Faculty of Health Sciences, University of Zaragoza, Zaragoza, Spain (D.J.-P., L.A.M.); Department of Education, University of Almería, Almería, Spain (E.G.A.); Department of Kinesiology, Iowa State University, Ames (D.-c.L.); Medical Research Council Epidemiology Unit, Institute of Metabolic Science, Addenbrooke’s Hospital, Cambridge, United Kingdom (V.E.-R.); Preventive Medicine Laboratory, Pennington Biomedical Research Center, Baton Rouge, LA (T.S.C.); and Department of Cardiovascular Diseases, John Ochsner Heart and Vascular Institute, Ochsner Clinical SchooleThe University of Queensland’s School of Medicine, New Orleans, LA (C.J.L.). Grant Support: This work was supported by the National Institutes of Health grants (grant nos. AG06945, HL62508, and R21DK088195) and in part by an unrestricted research grant from the Coca-Cola Company. Drs Jiménez-Pavón and Artero were supported by grants from the Spanish Ministry of Economy and Competitiveness (MINECO) (grant nos. RYC-2014-16938 and RYC-2014-16390, respectively). Correspondence: Address to David Jiménez-Pavón, PhD, GALENO Research Group, Department of Physical Education, University of Cádiz, Avd. República Saharaui s/n, 11519 Puerto Real, Cádiz, Spain ([email protected]).
REFERENCES 1. Chugh SS, Reinier K, Teodorescu C, et al. Epidemiology of sudden cardiac death: clinical and research implications. Prog Cardiovasc Dis. 2008;51(3):213-228. 2. Roger VL, Go AS, Lloyd-Jones DM, et al; American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statisticsd2012 update: a report from the American Heart Association [published correction appears in Circulation. 2012;125(22):e1002]. Circulation. 2012;125(1):e2-e220. 3. Mahmoud KD, de Smet BJ, Zijlstra F, Rihal CS, Holmes DR Jr. Sudden cardiac death: epidemiology, circadian variation, and triggers. Curr Probl Cardiol. 2011;36(2):56-80.
July 2016;91(7):849-857
n
http://dx.doi.org/10.1016/j.mayocp.2016.04.025 www.mayoclinicproceedings.org
CARDIORESPIRATORY FITNESS AND SUDDEN CARDIAC DEATH
4. Fox CS, Evans JC, Larson MG, Kannel WB, Levy D. Temporal trends in coronary heart disease mortality and sudden cardiac death from 1950 to 1999: the Framingham Heart Study. Circulation. 2004;110(5):522-527. 5. Adabag AS, Luepker RV, Roger VL, Gersh BJ. Sudden cardiac death: epidemiology and risk factors. Nat Rev Cardiol. 2010; 7(4):216-225. 6. Greenland P, Smith SC Jr, Grundy SM. Improving coronary heart disease risk assessment in asymptomatic people: role of traditional risk factors and noninvasive cardiovascular tests. Circulation. 2001;104(15):1863-1867. 7. Müller D, Agrawal R, Arntz HR. How sudden is sudden cardiac death? Circulation. 2006;114(11):1146-1150. 8. Huikuri HV, Castellanos A, Myerburg RJ. Sudden death due to cardiac arrhythmias. N Engl J Med. 2001;345(20):1473-1482. 9. Ekelund LG, Haskell WL, Johnson JL, Whaley FS, Criqui MH, Sheps DS. Physical fitness as a predictor of cardiovascular mortality in asymptomatic North American men: the lipid research clinics mortality follow-up study. N Engl J Med. 1988;319(21): 1379-1384. 10. Mora S, Redberg RF, Cui Y, et al. Ability of exercise testing to predict cardiovascular and all-cause death in asymptomatic women: a 20-year follow-up of the lipid research clinics prevalence study. JAMA. 2003;290(12):1600-1607. 11. Blair SN, Kohl HW III, Paffenbarger RS Jr, Clark DG, Cooper KH, Gibbons LW. Physical fitness and all-cause mortality: a prospective study of healthy men and women. JAMA. 1989;262(17):2395-2401. 12. Myers J, Prakash M, Froelicher V, Do D, Partington S, Atwood JE. Exercise capacity and mortality among men referred for exercise testing. N Engl J Med. 2002;346(11): 793-801. 13. Laukkanen JA, Mäkikallio TH, Rauramaa R, Kiviniemi V, Ronkainen K, Kurl S. Cardiorespiratory fitness is related to the risk of sudden cardiac death: a population-based followup study. J Am Coll Cardiol. 2010;56(18):1476-1483. 14. Laukkanen JA, Jennings JR, Kauhanen J, Mäkikallio TH, Ronkainen K, Kurl S. Relation of systemic blood pressure to sudden cardiac death. Am J Cardiol. 2012;110(3):378-382. 15. Laukkanen JA, Mäkikallio TH, Rauramaa R, Kurl S. Asymptomatic ST-segment depression during exercise testing and the risk of sudden cardiac death in middle-aged men: a population-based follow-up study. Eur Heart J. 2009;30(5): 558-565. 16. Chiuve SE, Fung TT, Rexrode KM, et al. Adherence to a lowrisk, healthy lifestyle and risk of sudden cardiac death among women. JAMA. 2011;306(1):62-69. 17. Lloyd-Jones DM, Hong Y, Labarthe D, et al; American Heart Association Strategic Planning Task Force and Statistics Committee. Defining and setting national goals for cardiovascular health promotion and disease reduction: The American Heart Association’s strategic Impact Goal through 2020 and beyond. Circulation. 2010;121(4):586-613. 18. Sui X, LaMonte MJ, Laditka JN, et al. Cardiorespiratory fitness and adiposity as mortality predictors in older adults. JAMA. 2007;298(21):2507-2516.
Mayo Clin Proc. n July 2016;91(7):849-857 www.mayoclinicproceedings.org
n
19. Pickering TG, Hall JE, Appel LJ, et al. Recommendations for blood pressure measurement in humans and experimental animals, part 1: blood pressure measurement in humans: a statement for professionals from the Subcommittee of Professional and Public Education of the American Heart Association Council on High Blood Pressure Research. Circulation. 2005;111(5): 697-716. 20. American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care. 2008;31(suppl 1):S55-S60. 21. Gibbons LW, Mitchell TL, Wei M, Blair SN, Cooper KH. Maximal exercise test as a predictor of risk for mortality from coronary heart disease in asymptomatic men. Am J Cardiol. 2000;86(1):53-58. 22. Balke B, Ware RW. An experimental study of physical fitness of Air Force personnel. U S Armed Forces Med J. 1959;10(6): 675-688. 23. Pollock ML, Bohannon RL, Cooper KH, et al. A comparative analysis of four protocols for maximal treadmill stress testing. Am Heart J. 1976;92(1):39-46. 24. Pollock ML, Foster C, Schmidt D, Hellman C, Linnerud AC, Ward A. Comparative analysis of physiologic responses to three different maximal graded exercise test protocols in healthy women. Am Heart J. 1982;103(3):363-373. 25. American College of Sport Medicine. ACSM’s Guidelines for Exercise Testing and Prescription. Philadelphia, PA: Lippincott Williams & Wilkins; 2005. 26. National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. Circulation. 2002;106(25):3143-3421. 27. Morris CK, Ueshima K, Kawaguchi T, Hideg A, Froelicher VF. The prognostic value of exercise capacity: a review of the literature. Am Heart J. 1991;122(5):1423-1431. 28. Jouven X, Empana JP, Schwartz PJ, Desnos M, Courbon D, Ducimetière P. Heart-rate profile during exercise as a predictor of sudden death. N Engl J Med. 2005;352(19): 1951-1958. 29. Tulppo MP, Mäkikallio TH, Seppänen T, Laukkanen RT, Huikuri HV. Vagal modulation of heart rate during exercise: effects of age and physical fitness. Am J Physiol. 1998;274(2, pt 2): H424-H429. 30. 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(1):47-53. 31. Mäkikallio TH, Huikuri HV, Mäkikallio A, et al. Prediction of sudden cardiac death by fractal analysis of heart rate variability in elderly subjects. J Am Coll Cardiol. 2001;37(5): 1395-1402. 32. Zheng ZJ, Croft JB, Giles WH, Mensah GA. Sudden cardiac death in the united states, 1989 to 1998. Circulation. 2001; 104(18):2158-2163.
http://dx.doi.org/10.1016/j.mayocp.2016.04.025
857
Cardiorespiratory Fitness and Risk of Sudden Cardiac Death in Men and Women in the United States: A Prospective Evaluation From the Aerobics Center Longitudinal Study David Jiménez-Pavón, PhD; Enrique G. Artero, PhD; Duck-chul Lee, PhD; Vanesa España-Romero, PhD; Xuemei Sui, MD, PhD; Russell R. Pate, PhD; Timothy S. Church, MD, PhD; Luis A. Moreno, MD, PhD; Carl J. Lavie, MD; and Steven N. Blair, PED Abstract Objectives: To examine the relation between cardiorespiratory fitness (CRF) and sudden cardiac death (SCD) in a large US adult population and to study the effects of hypertension, obesity, and health status on the relation of CRF with SCD. Patients and Methods: A total of 55,456 individuals (mean age, 44.2 years; 13,507 women) from the Aerobics Center Longitudinal Study, a prospective observational investigation (from January 2, 1974, through December 31, 2002), were included. Cardiorespiratory fitness was assessed by a maximal treadmill test, and baseline assessment included an extensive set of measurements. Results: There were 109 SCDs. An inverse risk of SCD was found across incremental CRF levels after adjusting for potential confounders. Participants with moderate and high CRF levels had 44% (hazard ratio, 0.56; 95% CI, 0.35-0.90) and 48% (hazard ratio, 0.52; 95% CI, 0.30-0.92) significantly lower risk of SCD, respectively, than did those with low CRF levels (P<.001). The risk of SCD decreased by 14% (hazard ratio, 0.86; 95% CI, 0.77-0.96) per 1-metabolic equivalent increase in the fully adjusted model. Hypertensive, overweight, or unhealthy individuals with moderate to high CRF levels had lower risks of SCD (ranging from 58% to 72% of lower risk) than did those with the same medical conditions and low CRF levels. Conclusion: The risk of SCD in US men and women could be partially reduced by ensuring moderate to high levels of CRF independently of other risk factors and especially in those who are hypertensive, overweight, or unhealthy. ª 2016 Mayo Foundation for Medical Education and Research
S
udden cardiac death (SCD) is recognized as a relevant cause of mortality, which in many cases occurs in the general population without previous symptoms of any coronary heart disease (CHD).1-3 Although 70% to 90% of the total incidence of SCD occurs in men, this cause of mortality has been documented in both sexes in individuals with a history of CHD or other major cardiovascular (CV) disease (CVD) risk factors and in those without a history of CHD or CVD.2,4 Therefore, SCD constitutes an important public health problem with multiple risk factors.5 Although numerous studies6-8 have attempted to identify those factors that are
n
Mayo Clin Proc. 2016;91(7):849-857
associated with higher SCD risk for the general population, conclusive evidence on this point remains somewhat elusive. A moderate to high level of cardiorespiratory fitness (CRF) has been associated with many CV benefits and has been established as a predictor of all-cause mortality in men and women.9-12 Moreover, the use of treadmill protocols and estimation of workloads based on treadmill speeds and inclines, typically reported as estimated metabolic equivalents (METs), as an indirect measurement of oxygen consumption, has widely reported to accurately predict these associations.9-12 However, only 1 study13 focused on the relation between
Mayo Clin Proc. n July 2016;91(7):849-857 n http://dx.doi.org/10.1016/j.mayocp.2016.04.025 www.mayoclinicproceedings.org n ª 2016 Mayo Foundation for Medical Education and Research
From the GALENO Research Group, Department of Physical Education, University of Cádiz, Puerto Real, Cádiz, Spain (D.J.-P., V.E.-R.); Department of Exercise Science (D.J.-P., E.G.A., X.S., R.R.P., S.N.B.) and Department of Epidemiology and Biostatistics (S.N.B.), University of South Carolina, Columbia; GENUD (Growth, Exercise, Affiliations continued at the end of this article.
849
MAYO CLINIC PROCEEDINGS
CRF and SCD, in which authors studied 2368 European middle-aged men and reported that CRF (cycle ergometer test) was inversely related to the risk of SCD. In addition, other authors studied the effect of specific risk factors, such as systemic blood pressure (BP) and hypertension (HTN)14 and asymptomatic ST-segment depression15 in men or a cluster of lifestyle factors in women,16 on the risk of SCD. One of the American Heart Association’s 2020 Impact Goals17 is to decrease CVD mortality. Thus, prevention of SCD by detecting early risk factors could be a useful tool contributing to the American Heart Association’s goal through decreasing the incidence of SCD at the population level. The identification of these risk factors would facilitate the large-scale screening of those participants at higher risk of SCD. However, to our knowledge, there has been no large-scale prospective assessment of the relation between CRF and SCD, including men and women, and assessment of the effect of CRF, especially combined with the other important CVD risk factors. Therefore, the present study aimed to examine the relation between CRF and SCD in a large population of men and women in the United States. We also sought to study the specific effects of HTN, obesity, and overall health status on the relation of CRF with SCD. PATIENTS AND METHODS The present report is based on data from the Aerobics Center Longitudinal Study, a prospective observational investigation into the association of clinical and lifestyle factors with disease and health outcomes in patients examined at the Cooper Clinic in Dallas, Texas.11,17 Participants came to the clinic for periodic preventive health examinations and for counseling on diet, exercise, and other lifestyle factors associated with an increased risk of chronic disease. Participants were volunteers, not paid, and were not recruited into the study, but sent by their employers or physicians or were self-referred for the examination. The Cooper Institute’s institutional review board reviewed and approved the study protocol annually. For the present analysis, participants 20 years and older with complete data on CRF and a group of potential confounders including 850
Mayo Clin Proc.
n
health behaviors, blood parameters, and disease-related information were included. From a total of 59,611 participants, we excluded those who failed to achieve at least the 85% of age-predicted maximum heart rate during the treadmill test (n¼1868) or were followed for less than 1 year (n¼2287). These criteria resulted in 55,456 individuals (13,507 women) with a mean age of 44.2 years (ranging from 20 to 100 years), whose baseline examination occurred from January 2, 1974, through December 31, 2003. Participants were predominantly white, well-educated, and belonged to the middle and upper socioeconomic strata. Outcomes and Follow-Up Procedures of the clinical examinations are described in detail elsewhere.11,18 Briefly, the baseline assessment was performed after an overnight fast and included an extensive physical examination and an array of clinical measurements. Body mass index (BMI) was calculated as the weight in kilograms divided by the height in meters squared (kg/m2) and measured using a standard clinical scale and stadiometer. Overweight and obesity were defined as a BMI of 25 to 29.9 and 30 or more kg/m2, respectively. Resting systolic and diastolic BP was measured in the seated position as the first and fifth Korotkoff sounds by using standard auscultation methods after at least 5 minutes of sitting quietly and recorded as the average of at least 2 readings separately by 2 minutes.19 If the first 2 readings differed by more than 5 mm Hg, additional readings were obtained and averaged; HTN was defined as systolic or diastolic BP of 140/90 mm Hg or greater or a history of physician diagnosed HTN. Concentrations of total cholesterol and fasting glucose were measured using automated techniques in accordance with the standards of the Centers for Disease Control and Prevention lipid standardization program. Hypercholesterolemia was defined as a serum total cholesterol level of 240 mg/dL or greater (to convert to mmol/L, multiply by 0.0259). Diabetes mellitus was defined according to the American Diabetes Association criteria,20 that is, a fasting plasma glucose level of 126 mg/dL or greater, use of insulin, or self-report of previous physician diagnosis. Participants completed a standardized questionnaire on medical history that
July 2016;91(7):849-857
n
http://dx.doi.org/10.1016/j.mayocp.2016.04.025 www.mayoclinicproceedings.org
CARDIORESPIRATORY FITNESS AND SUDDEN CARDIAC DEATH
included a personal history of myocardial infarction, stroke, HTN, diabetes mellitus, and cancer; a family history of CVD; smoking status (never, former, or current smoker); alcohol intake (heavy drinker or not, defined as >14 and 7 drinks/wk for men and women, respectively; 1 unit of alcohol was defined as 12 oz of beer, 5 oz of wine, or 1.5 oz of hard liquor); and physical activity (active or inactive). Physical inactivity was defined as reporting no physical activity during leisure time in the 3 months before the baseline examination. Electrocardiography showing abnormalities (both at rest and exercise) were largely defined as rhythm and conduction disturbances and ischemic ST-T wave abnormalities, as described elsewhere.21 Cardiorespiratory Fitness. Cardiorespiratory fitness was assessed by a maximum treadmill exercise test using a modified Balke protocol.11,22 Participants began walking at a speed of 88 m/min without elevation. After the first minute, the elevation was increased by 2%; thereafter, the elevation was increased by 1% per minute until the 25th minute. After 25 minutes, the elevation did not change but the speed was increased by 5.4 m/min each minute until the test end point. Participants were encouraged to give maximal effort, and the test was stopped when they reached volition exhaustion or the doctor intervened for medical reasons. The mean ageepredicted maximum heart rate (220age) achieved during exercise was used to determine individuals who achieved maximal effort. Exercise duration using this protocol is highly correlated with measured maximum oxygen uptake (r0.90) in both men23 and women.24 To standardize the interpretation of exercise test performance, maximal METs (1 MET¼3.5 mL oxygen uptake/kg per minute) were estimated on the basis of the final treadmill speed and grade.25 Cardiorespiratory fitness was categorized as low, moderate, or high age- and sex-specific distributions of maximal exercise duration from the entire Aerobics Center Longitudinal Study population according to previous publications.11,18 Sudden Cardiac Death. All participants were followed from the date of their baseline examination until their date of death or until Mayo Clin Proc. n July 2016;91(7):849-857 www.mayoclinicproceedings.org
n
December 31, 2003. We computed personyears of exposure as the sum of follow-up time in decedents and survivors. The National Death Index was the primary data source for mortality surveillance, augmented with death certificates as previously used.13 The National Death Index is an accurate method of ascertaining deaths in observational studies, with high sensitivity (96%) and specificity (100%).25 The underlying and contributing cause of death was determined from the index or by a nosologist’s review of official death certificates obtained from the department of vital records in the decedent’s state of residence. Sudden cardiac death was defined by the International Classification of Diseases, Ninth Revision code 798.1 before 1999 and by the International Classification of Diseases, Tenth Revision code I46 between 1999 and 2003 according to others.13 Statistical Analyses Descriptive analyses summarized baseline characteristics of the participants according to CRF categories. Tests for linear trends across CRF categories were performed using general linear models for continuous variables and the chi-square test for categorical variables. Cox proportional hazards regression was used to estimate hazard ratios (HRs), 95% CIs, and mortality rates (deaths per 10,000 person-years of follow-up) according to CRF categories. No significant effect modification by sex was observed on SCD using interaction terms (CRFsex) in Cox regressions; thus, the analyses were performed on the pooled sample. Multivariable analyses according to 3 models of adjustments were performed: model 1 was the unadjusted model; model 2 included adjustments for age and sex; and model 3 included adjustments for model 2 plus examination year, BMI, current smoking, heavy alcohol intake (>14 or 7 drinks/wk for men and women, respectively), diabetes mellitus (presence or absence), HTN (presence or absence), electrocardiogram showing abnormalities at rest or exercise (presence or absence), and parental history of CVD. Further analyses were performed to estimate the hazards for SCD by aerobic capacity using METs as a continuous variable and using the same models of adjustments.
http://dx.doi.org/10.1016/j.mayocp.2016.04.025
851
MAYO CLINIC PROCEEDINGS
In addition, the effects of HTN, obesity, and health status (all presence or absence) on the association of CRF with risk of SCD were explored. Health status was defined as the presence or absence of electrocardiogram showing abnormalities and personal history of CVD or cancer. The analyses were adjusted for age, sex, examination year, and electrocardiogram showing abnormalities (only for the presence or absence of HTN) according to the number of SCDs per group. Further analyses were performed to estimate the effect of health status on the hazards for SCD by METs.
RESULTS Descriptive Characteristics There were 109 SCDs (20 per 10,000 personyears) in men and women during an average follow-up of 14.7 years. Table 1 summarizes the baseline characteristics of the participants according to CRF categories. Body mass index, total cholesterol level, fasting blood glucose level, and BP were lower in those with higher CRF levels. The prevalence of participants who were inactive, current smokers, and having hypercholesterolemia, diabetes mellitus, HTN, electrocardiogram showing abnormalities,
TABLE 1. Baseline Characteristics of the Participants from the Aerobics Center Longitudinal Study (1974-2003) According to the Cardiorespiratory Fitness Levels and Sudden Cardiac Deatha,b Cardiorespiratory fitness level Characteristic
All (N¼55,456)
Low (n¼8404)
Moderate (n¼21,458)
High (n¼25,594)
P value for linear trendc
Age (y) Body mass index (kg/m2) Treadmill time (min) Maximum METs Total cholesterol level (mg/dL)d Fasting blood glucose level (mg/dL) Waist circumference (cm) Triglyceride level (mg/dL) Blood pressure (mm Hg) Systolic Diastolic Physically inactivee Current smokers Heavy drinkersf Hypercholesterolemiag Diabetes mellitush HTNi Electrocardiogram showing abnormalitiesj Personal history of CVD Cancerk Parental history of CVD Metabolic syndromel
44.210 25.74.1 16.85.2 11.12.5 206.239.7 98.816.8 87.219.6 128.4104.7
43.59.4 29.05.6 10.52.9 8.21.4 216.041.9 103.825.9 98.721.6 175.8141.9
44.39.9 26.23.7 14.93.1 10.21.4 209.839.8 99.216.3 90.118.8 139.3102.4
44.210.2 24.23.0 20.54.3 12.92.2 199.937.8 96.712.6 82.218.0 103.683.0
<.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001
119.714.4 80.09.9 16,019 (28.9) 8946 (16.1) 9801 (17.7) 14,938 (26.9) 2982 (5.4) 15,807 (28.5) 4237 (7.6) 658 (1.2) 1805 (3.3) 14,862 (26.8) 8964 (16.2)
123.314.9 83.010.3 5243 (62.4) 2401 (28.6) 1472 (17.5) 2853 (33.9) 834 (9.9) 3503 (41.7) 874 (10.4) 192 (2.3) 156 (1.9) 2461 (29.3) 2512 (29.9)
120.014.2 80.79.9 7915 (36.9) 4145 (19.3) 3771 (17.6) 6329 (29.5) 1225 (5.7) 6620 (30.9) 1719 (8.0) 282 (1.3) 626 (2.9) 5831 (27.2) 4354 (20.3)
118.314.1 78.69.4 2861 (11.2) 2400 (9.4) 4558 (17.8) 5756 (22.5) 923 (3.6) 5684 (22.2) 1644 (6.4) 184 (0.7) 1023 (4.0) 6570 (25.7) 2098 (8.2)
<.001 <.001 <.001 <.001 .74 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001
CVD ¼ cardiovascular disease; HTN ¼ hypertension. Data are presented as mean SD or as No. (percentage). c Differences across cardiorespiratory fitness categories using general linear models for continuous variables or the c2 test for categorical variables. d SI conversion factor: To convert mg/dL values to mmol/L, multiply by 0.0259. e Defined as reporting no physical activity during leisure time in the 3 mo before the examination. f Defined as >14 and 7 drinks/wk for men and women, respectively (1 unit of alcohol defined as 12 oz of beer, 5 oz of wine, and 1.5 oz of hard liquor). g Defined as total cholesterol level of 240 mg/dL or previous physician diagnosis. h Defined as fasting plasma glucose level of 126 mg/dL, previous physician diagnosis, or use of insulin. i Defined as resting blood pressure of 140/90 mm Hg or previous physician diagnosis. j Defined as rhythm and conduction disturbances and ischemic ST-T wave abnormalities (at rest or exercise). k Defined as personal history of cancer. l Metabolic syndrome established using the National Cholesterol Education Program Adult Treatment Panel III criteria and was based on the presence of 3 risk factors.26 a
b
852
Mayo Clin Proc.
n
July 2016;91(7):849-857
n
http://dx.doi.org/10.1016/j.mayocp.2016.04.025 www.mayoclinicproceedings.org
CARDIORESPIRATORY FITNESS AND SUDDEN CARDIAC DEATH
and parental history of CVD was lower at higher CRF levels. When corrections for multiple comparisons were applied, the results with P values less than .002 remained significant. Hazard Ratios for SCD by CRF Levels Table 2 lists the mortality rate and HR for SCD calculated using 3 models of adjustments. Mortality rates for SCD were lower at higher CRF levels. An inverse risk of SCD was found across incremental CRF levels in all the models (P value for linear trend: P<.05 for model 1 and P<.001 for models 2 and 3). Participants with moderate and high CRF levels had 44% (HR, 0.56; 95% CI, 0.35-0.90) and 48% (HR, 0.52; 95% CI, 0.30-0.92) significantly lower risk of SCD, respectively, than did those with low CRF levels in the multivariate-adjusted model (model 3). The risk of SCD decreased by 14% (HR, 0.86; 95% CI, 0.77-0.96) per 1-MET increase in the fully adjusted model (model 3). Effects of HTN, Obesity, and Health Status The Figure shows the risk of SCD across CRF levels by the presence or absence of HTN (panel A), by the presence or absence of normal weight or overweight or obese (panel B), and in healthy or unhealthy groups (panel C) after adjustments for age, sex, examination year, and electrocardiogram showing abnormalities (only for the presence or absence of HTN). Inverse trends in the risk of SCD across CRF levels were observed in all subgroups, but only reaching statistical significance in HTN (and the normotensive group with high CRF
levels), overweight or obese, and healthy or unhealthy (moderate CRF levels) groups. Those participants identified with HTN and with moderate or high CRF levels had 65% (HR, 0.35; 95% CI, 0.18e0.67) and 72% (HR, 0.28; 95% CI, 0.12e0.61) lower risk of SCD, respectively, than did those with low CRF levels. The normotensive participants with high CRF levels had 51% (HR, 0.49; 95% CI, 0.24e0.98) lower risk of SCD than did those with low CRF levels. Similarly, the risk of SCD was lower in overweight or obese participants with moderate or high CRF levels (HR, 0.42; 95% CI, 0.25e0.72 and HR, 0.36; 95% CI, 0.18e0.71, respectively). The healthy group had a mean CRF level of 11 METs vs the unhealthy group with 10 METs (P<.001). In the unhealthy group, those participants with moderate CRF levels had 69% (HR, 0.31; 95% CI, 0.13e0.72) lower risk of SCD, whereas participants classified in the healthy group and with moderate or high CRF levels had 47% (HR, 0.53; 95% CI, 0.31e0.90) and 65% (HR, 0.35; 95% CI, 0.19e0.65), respectively, lower risk of SCD than did those with low CRF levels. Each 1-MET increase was related to 20% (HR, 0.80; 95% CI, 0.71e0.90; P¼.001) lower risk in the healthy group and to 18% (HR, 0.82; 95% CI, 0.68e0.98; P¼.05) lower risk in the unhealthy group. Sensitivity Analyses The results were similar when all the analyses were repeated using an older age group (age, 40 years; n¼36,720); specifically, CRF
TABLE 2. Rates and Hazards Ratios for SCD in Men and Women According to Cardiorespiratory Fitness Levelsa
Variable Cardiorespiratory fitness level Low Moderate High P value for linear trend Per 1-MET increase
No. of SCDs (n)
Mortality rate (%)b
38 (8366) 41 (21,417) 30 (25,564)
2.6 1.2 0.9
Hazard ratio (95% CI) Model 1
Model 2
Model 3
1.00 (reference) 0.62 (0.40-0.96) 0.56 (0.35-0.91) .03 0.77 (0.70-0.84)
1.00 (reference) 0.52 (0.33-0.80) 0.40 (0.25-0.66) <.001 0.82 (0.75-0.90)
1.00 (reference) 0.56 (0.35-0.90) 0.52 (0.30-0.92) <.001 0.86 (0.77-0.96)
MET ¼ metabolic equivalent; n ¼ number of participants; SCD ¼ sudden cardiac death. Per 10,000 person-years, adjusted for age, sex, and examination year. Model 1 was unadjusted; model 2 was adjusted for age and sex; model 3 was adjusted for model 2 plus examination year, body mass index, current smoking, heavy alcohol intake (>14 and 7 drinks/wk for men and women, respectively), diabetes mellitus (presence or absence), hypertension (presence or absence), electrocardiogram showing abnormalities (presence or absence), and parental history of cardiovascular disease.
a
b
Mayo Clin Proc. n July 2016;91(7):849-857 www.mayoclinicproceedings.org
n
http://dx.doi.org/10.1016/j.mayocp.2016.04.025
853
MAYO CLINIC PROCEEDINGS
mortality rates were 3.9%, 1.9%, and 1.3% for low, middle, and high CRF levels, respectively. Alternatively, Cox regression models using personal history of CVD instead of electrocardiogram showing abnormalities did not modify the main results. Furthermore, including those who did not reach 85% of the age-predicted maximum heart rate did not significantly change the main findings. Moreover, the incidence of SCD (reported as a percentage of allcause mortality) across fitness levels and main risk factors was similar to the incidence of SCD in the overall population (data not shown). Finally, the results were not materially altered when analyses were repeated excluding participants without SCD and including those with other causes of mortality.
Hazard ratio of sudden cardiac death
2.5
2.0
1.5 Reference
Reference
1.0 0.62
0.5
0.49 0.35
0.28
0.0 Low
Moderate High CRF No. of deaths 15 26 21 (No. of participants) (4901) (14,839) (19,910)
A
Low
Moderate High CRF 15 9 23 (3503) (6621) (5684)
Normotensive
Hypertensive
Hazard ratio of sudden cardiac death
2.5
2.0
1.5 Reference
Reference
1.0 0.57
0.5
0.42 0.42
0.36
0.0 Low
Moderate High CRF No. of deaths 7 15 17 (No. of participants) (1973) (8075) (15,932)
B
Normal weight
Low
Moderate High CRF 31 26 13 (6431) (13,385) (9662) Overweight and obese
Hazard ratio of sudden cardiac death
2.5
2.0
1.5 Reference
Reference
1.0
0.53
0.5
0.35
0.31
0.41
0.0 Low
Moderate High CRF 25 32 21 No. of deaths (No. of participants) (7223) (19,046) (22,940)
C
Healthy
Low
Moderate High CRF 13 9 9 (1081) (2414) (2654) Unhealthy
FIGURE. Association of cardiorespiratory fitness (CRF) (low, moderate, and high) with risk of sudden cardiac death (A) by the presence or absence of hypertension, (B) by the presence or absence of normal weight or overweight or obesity, and (C) in healthy or unhealthy groups, in which unhealthy was defined as participant having electrocardiogram showing abnormalities or personal history of cardiovascular disease or cancer. The analyses were adjusted for age, sex, examination year, and electrocardiogram showing abnormalities (only for the presence or absence of hypertension) according to the number of sudden cardiac deaths per group.
854
Mayo Clin Proc.
n
DISCUSSION The main finding from these prospective analyses was that participants with moderate to high CRF levels had about half the risk of SCD than did individuals with low CRF levels, and this was independent of other potentially confounding factors. Therefore, low CRF levels is a major risk factor for SCD in a large and general population of men and women in the United States This study also found that the decreases in the risk of SCD by having moderate to high CRF levels were even higher (ranging from 58% to 72% of lower risk) in the HTN, overweight or obese, and unhealthy participants. Each 1-MET increase in the level of CRF corresponded to an adjusted risk reduction in both healthy and unhealthy populations (20% and 18%, respectively). The only previous study addressing the relation between CRF and SCD was performed by Laukkanen et al,13 in which 2368 Finish men (age, 42-60 years) were studied during 17 years of follow-up by using a cycle ergometer test for the estimation of CRF (in METs). These authors13 found that each 1-MET increase in CRF levels was associated with a reduction in the risk of SCD of 22% in an adjusted risk model. Our study adds to this previous research by analyzing a larger population including both men and women from the United States and using a maximal treadmill exercise test for the estimation of CRF, which is the most used means of assessing exercise capacity in the United States and
July 2016;91(7):849-857
n
http://dx.doi.org/10.1016/j.mayocp.2016.04.025 www.mayoclinicproceedings.org
CARDIORESPIRATORY FITNESS AND SUDDEN CARDIAC DEATH
also recreates the natural skill of walking/ running as a tool for population-based screenings. Our results confirmed the previous findings and extend them to a large sample of US men and women, highlighting the importance of having moderate to high levels of CRF (10.2 and 12.9 METs, respectively) in relation to SCD. Other authors14-16 have also assessed SCD in relation to other independent or clustered risk factors. They reported that men with HTN (systolic BP >145 mm Hg; n¼2666) had a 2.04-fold higher adjusted risk for SCD than did those with systolic BP less than 123 mm Hg.14 Moreover, Laukkanen et al15 studied the association between silent ST-segment depression during and after the maximal exercise test and the risk of SCD in men (n¼1769) and found that ST-segment depression during and after the exercise test was a strong predictor of the risk of SCD, with special relevance in smokers, individuals with hypercholesterolemia, and men with HTN. Recently, Chiuve et al16 performed a prospective study with 81,722 American women during 26 years of follow-up, analyzing the degree to which adherence to a healthy lifestyle (as a cluster of 4 self-reported components: smoking, physical activity, diet, and weight) affected the risk of SCD, and found that all 4 low-risk factors were significantly and independently related to a lower risk of SCD in women. In our present study, the particular effects of other risk factors with regard to the relation of CRF and SCD were also assessed. We confirmed the relevance of these risk factors as reported in previous studies,14,16 but also provided a new approach from the viewpoint of CRF levels. In fact, we observed that CRF levels were related to a lower risk of SCD in both healthy and unhealthy men and women, but these relations were stronger in those participants with HTN, who were overweight or obese, and who were unhealthy. These results confirmed the previous findings on the relation of HTN, BMI, and healthy status with SCD,14,16 but also suggested that higher CRF levels may add additional strength to the associations with SCD in men and women with, but also without, CVD symptoms or risk factors. The understanding of the possible mechanism underlying the association of CRF with risk of SCD is not clear, and it cannot be Mayo Clin Proc. n July 2016;91(7):849-857 www.mayoclinicproceedings.org
n
confirmed in the present study. However, we hypothesize that similar reasons for the beneficial association of CRF with CHD-related mortality would also provide protection from SCD. In this regard, having a moderate or high level of CRF has been repeatedly shown to be associated with a lower risk of CVD morbidity and mortality in apparently healthy individuals, as well as in those with other CVD risk factors.9,11,12,21,27 Moreover, an important report by Jouven et al28 highlighted that alterations in the neural control of cardiac function contribute to the risk of SCD, and they reported that abnormal heart rate profiles during exercise and recovery were related to a higher risk of SCD.29-31 The CRF level has been found to affect the vagal modulation of heart rate as well as to positively affect the CV autonomic function. Consequently, having a moderate to high level of CRF could decrease the risk of SCD through the different benefits on the CV and nervous systems attributed to exercise and overall physical fitness. There are several limitations of our study that should be considered. The Aerobics Center Longitudinal Study study sample was mainly composed of white men and women; thus, as black populations have higher mortality rates for SCD than do whites or others,32 prospective studies on populations from different ethnicities are needed. Another limitation is that a detailed history on medication usage was not available. However, because the present study does not have a randomized controlled trial design, we cannot establish the dose-response effect of CRF increments on SCD. No information on the causes of SCD was reported, such as congenital heart disease or electrical or structural abnormalities of the heart and great vessels. Moreover, murmur was not included in the database, nor were echocardiograms routinely performed. However, because of the long follow-up necessary for this kind of study and the large scale of the sample size required, it would be extremely difficult to corroborate this hypothesis or perform more detailed assessments in a randomized study. Because previous publications reported associations of CRF with other mortality outcomes using this data set, but SCD was not explored as an outcome, we think these findings might be useful to understand the relation between
http://dx.doi.org/10.1016/j.mayocp.2016.04.025
855
MAYO CLINIC PROCEEDINGS
CRF and SCD. Finally, although our sample size (>55,000 and w25% women) and follow-up period (average, 14.7 years) were quite robust, the incidence of SCD was rare, and our analysis of only 109 SCD events somewhat limits the overall statistical power of our study. This relatively low event rate of SCD observed could be partly explained by the outcome classification. A death is classified as SCD when it occurred within 24 hours of the onset of symptoms (including nonwitnessed SCD cases as well) and there are not any other noncardiac causes of sudden death. If a death was witnessed, an unexpected cardiac event that occurs within 1 hour from the onset of symptoms is typically classified as SCD. In this sense, the use of International Classification of Diseases codes to identify SCD could have underestimated the overall number of deaths and all SCD cases may not have been included; however, the National Death Index has shown to be an accurate method of ascertaining deaths in observational studies25 and was previously used.13 Finally, the indirect use of CRF, as there was no exchange of gases during the test, could be another limitation; however, this indirect use of CRF has shown to be sensitive in previous studies.11,18 The present study also has several additional strengths, including the use of highly reliable measurement through a maximal treadmill exercise protocol to determine CRF levels, which allowed a better transference to habitual activities (walking or running) in comparison with cycling13 and to avoid potential bias derived from the use of a questionnaire.16 CONCLUSION The risk of SCD in US men and women is significantly lower in those having moderate to high levels of CRF independently of other risk factors and even in those who are asymptomatic. To highlight the importance of having a higher CRF level is the fact that in particularly vulnerable groups, such as hypertensive, obese, or unhealthy populations, the reduction in the risk of SCD when having moderate to high CRF levels could achieve the approximately 75% of the total risk. Therefore, not only is obtaining the CRF level a useful tool in screening for the risk of CVD and CVD mortality, it also provides additional information on the 856
Mayo Clin Proc.
n
risk of SCD in asymptomatic men and women. Furthermore, our data provide information on the potential benefits of moderate to high CRF levels, especially in higher risk patients, for the primary prevention of SCD.
ACKNOWLEDGMENTS We thank physicians and technicians at the Cooper Clinic for collecting the data and staff at the Cooper Institute for data entry and data management. Abbreviations and Acronyms: BMI = body mass index; BP = blood pressure; CHD = coronary heart disease; CRF = cardiorespiratory fitness; CV = cardiovascular; CVD = cardiovascular disease; HR = hazard ratio; HTN = hypertension; MET = metabolic equivalent; SCD = sudden cardiac death Affiliations (Continued from the first page of this article.): Nutrition and Development) Research Group, Faculty of Health Sciences, University of Zaragoza, Zaragoza, Spain (D.J.-P., L.A.M.); Department of Education, University of Almería, Almería, Spain (E.G.A.); Department of Kinesiology, Iowa State University, Ames (D.-c.L.); Medical Research Council Epidemiology Unit, Institute of Metabolic Science, Addenbrooke’s Hospital, Cambridge, United Kingdom (V.E.-R.); Preventive Medicine Laboratory, Pennington Biomedical Research Center, Baton Rouge, LA (T.S.C.); and Department of Cardiovascular Diseases, John Ochsner Heart and Vascular Institute, Ochsner Clinical SchooleThe University of Queensland’s School of Medicine, New Orleans, LA (C.J.L.). Grant Support: This work was supported by the National Institutes of Health grants (grant nos. AG06945, HL62508, and R21DK088195) and in part by an unrestricted research grant from the Coca-Cola Company. Drs Jiménez-Pavón and Artero were supported by grants from the Spanish Ministry of Economy and Competitiveness (MINECO) (grant nos. RYC-2014-16938 and RYC-2014-16390, respectively). Correspondence: Address to David Jiménez-Pavón, PhD, GALENO Research Group, Department of Physical Education, University of Cádiz, Avd. República Saharaui s/n, 11519 Puerto Real, Cádiz, Spain ([email protected]).
REFERENCES 1. Chugh SS, Reinier K, Teodorescu C, et al. Epidemiology of sudden cardiac death: clinical and research implications. Prog Cardiovasc Dis. 2008;51(3):213-228. 2. Roger VL, Go AS, Lloyd-Jones DM, et al; American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statisticsd2012 update: a report from the American Heart Association [published correction appears in Circulation. 2012;125(22):e1002]. Circulation. 2012;125(1):e2-e220. 3. Mahmoud KD, de Smet BJ, Zijlstra F, Rihal CS, Holmes DR Jr. Sudden cardiac death: epidemiology, circadian variation, and triggers. Curr Probl Cardiol. 2011;36(2):56-80.
July 2016;91(7):849-857
n
http://dx.doi.org/10.1016/j.mayocp.2016.04.025 www.mayoclinicproceedings.org
CARDIORESPIRATORY FITNESS AND SUDDEN CARDIAC DEATH
4. Fox CS, Evans JC, Larson MG, Kannel WB, Levy D. Temporal trends in coronary heart disease mortality and sudden cardiac death from 1950 to 1999: the Framingham Heart Study. Circulation. 2004;110(5):522-527. 5. Adabag AS, Luepker RV, Roger VL, Gersh BJ. Sudden cardiac death: epidemiology and risk factors. Nat Rev Cardiol. 2010; 7(4):216-225. 6. Greenland P, Smith SC Jr, Grundy SM. Improving coronary heart disease risk assessment in asymptomatic people: role of traditional risk factors and noninvasive cardiovascular tests. Circulation. 2001;104(15):1863-1867. 7. Müller D, Agrawal R, Arntz HR. How sudden is sudden cardiac death? Circulation. 2006;114(11):1146-1150. 8. Huikuri HV, Castellanos A, Myerburg RJ. Sudden death due to cardiac arrhythmias. N Engl J Med. 2001;345(20):1473-1482. 9. Ekelund LG, Haskell WL, Johnson JL, Whaley FS, Criqui MH, Sheps DS. Physical fitness as a predictor of cardiovascular mortality in asymptomatic North American men: the lipid research clinics mortality follow-up study. N Engl J Med. 1988;319(21): 1379-1384. 10. Mora S, Redberg RF, Cui Y, et al. Ability of exercise testing to predict cardiovascular and all-cause death in asymptomatic women: a 20-year follow-up of the lipid research clinics prevalence study. JAMA. 2003;290(12):1600-1607. 11. Blair SN, Kohl HW III, Paffenbarger RS Jr, Clark DG, Cooper KH, Gibbons LW. Physical fitness and all-cause mortality: a prospective study of healthy men and women. JAMA. 1989;262(17):2395-2401. 12. Myers J, Prakash M, Froelicher V, Do D, Partington S, Atwood JE. Exercise capacity and mortality among men referred for exercise testing. N Engl J Med. 2002;346(11): 793-801. 13. Laukkanen JA, Mäkikallio TH, Rauramaa R, Kiviniemi V, Ronkainen K, Kurl S. Cardiorespiratory fitness is related to the risk of sudden cardiac death: a population-based followup study. J Am Coll Cardiol. 2010;56(18):1476-1483. 14. Laukkanen JA, Jennings JR, Kauhanen J, Mäkikallio TH, Ronkainen K, Kurl S. Relation of systemic blood pressure to sudden cardiac death. Am J Cardiol. 2012;110(3):378-382. 15. Laukkanen JA, Mäkikallio TH, Rauramaa R, Kurl S. Asymptomatic ST-segment depression during exercise testing and the risk of sudden cardiac death in middle-aged men: a population-based follow-up study. Eur Heart J. 2009;30(5): 558-565. 16. Chiuve SE, Fung TT, Rexrode KM, et al. Adherence to a lowrisk, healthy lifestyle and risk of sudden cardiac death among women. JAMA. 2011;306(1):62-69. 17. Lloyd-Jones DM, Hong Y, Labarthe D, et al; American Heart Association Strategic Planning Task Force and Statistics Committee. Defining and setting national goals for cardiovascular health promotion and disease reduction: The American Heart Association’s strategic Impact Goal through 2020 and beyond. Circulation. 2010;121(4):586-613. 18. Sui X, LaMonte MJ, Laditka JN, et al. Cardiorespiratory fitness and adiposity as mortality predictors in older adults. JAMA. 2007;298(21):2507-2516.
Mayo Clin Proc. n July 2016;91(7):849-857 www.mayoclinicproceedings.org
n
19. Pickering TG, Hall JE, Appel LJ, et al. Recommendations for blood pressure measurement in humans and experimental animals, part 1: blood pressure measurement in humans: a statement for professionals from the Subcommittee of Professional and Public Education of the American Heart Association Council on High Blood Pressure Research. Circulation. 2005;111(5): 697-716. 20. American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care. 2008;31(suppl 1):S55-S60. 21. Gibbons LW, Mitchell TL, Wei M, Blair SN, Cooper KH. Maximal exercise test as a predictor of risk for mortality from coronary heart disease in asymptomatic men. Am J Cardiol. 2000;86(1):53-58. 22. Balke B, Ware RW. An experimental study of physical fitness of Air Force personnel. U S Armed Forces Med J. 1959;10(6): 675-688. 23. Pollock ML, Bohannon RL, Cooper KH, et al. A comparative analysis of four protocols for maximal treadmill stress testing. Am Heart J. 1976;92(1):39-46. 24. Pollock ML, Foster C, Schmidt D, Hellman C, Linnerud AC, Ward A. Comparative analysis of physiologic responses to three different maximal graded exercise test protocols in healthy women. Am Heart J. 1982;103(3):363-373. 25. American College of Sport Medicine. ACSM’s Guidelines for Exercise Testing and Prescription. Philadelphia, PA: Lippincott Williams & Wilkins; 2005. 26. National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. Circulation. 2002;106(25):3143-3421. 27. Morris CK, Ueshima K, Kawaguchi T, Hideg A, Froelicher VF. The prognostic value of exercise capacity: a review of the literature. Am Heart J. 1991;122(5):1423-1431. 28. Jouven X, Empana JP, Schwartz PJ, Desnos M, Courbon D, Ducimetière P. Heart-rate profile during exercise as a predictor of sudden death. N Engl J Med. 2005;352(19): 1951-1958. 29. Tulppo MP, Mäkikallio TH, Seppänen T, Laukkanen RT, Huikuri HV. Vagal modulation of heart rate during exercise: effects of age and physical fitness. Am J Physiol. 1998;274(2, pt 2): H424-H429. 30. 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(1):47-53. 31. Mäkikallio TH, Huikuri HV, Mäkikallio A, et al. Prediction of sudden cardiac death by fractal analysis of heart rate variability in elderly subjects. J Am Coll Cardiol. 2001;37(5): 1395-1402. 32. Zheng ZJ, Croft JB, Giles WH, Mensah GA. Sudden cardiac death in the united states, 1989 to 1998. Circulation. 2001; 104(18):2158-2163.
http://dx.doi.org/10.1016/j.mayocp.2016.04.025
857