Share this article on:

High Cardiorespiratory Fitness Levels Slow the Decline in Peak Heart Rate with Age

OZEMEK, CEMAL1; WHALEY, MITCHELL H.1; FINCH, W. HOLMES2; KAMINSKY, LEONARD A.1

Medicine & Science in Sports & Exercise: January 2016 - Volume 48 - Issue 1 - p 73–81
doi: 10.1249/MSS.0000000000000745
Epidemiology

Introduction: Although it is well accepted that peak HR (HRpeak) decreases with age, there is no clear consensus on the rate of this decline and the influence of cardiorespiratory fitness (CRF) on the rate of decline.

Methods: Treadmill cardiopulmonary exercise test (CPET) results with RER ≥ 1.0 from participants (1849 men and 1469 women; 18–80 yr) of a university-based health assessment/fitness program were retrospectively examined using cross-sectional and longitudinal study designs. All subjects were free of overt cardiovascular disease and were not taking HR-altering medications. Only subjects having completed ≥2 cardiopulmonary exercise tests with ≥1 yr between test dates were included in the longitudinal analysis (418 men and 225 women). Subjects were categorized into CRF categories (high, moderate, and low) relative to age and gender normative classifications. A general linear-model univariate analysis was performed to test the effect of CRF on the decline in HRpeak with age.

Results: In both cross-sectional and longitudinal analyses, HRpeak declined at a significantly faster rate across the CRF categories (cross-sectional: −0.60, −0.78, and −0.87 bpm·yr−1, respectively; longitudinal: −0.61, −0.82, and −1.02 bpm·yr−1, respectively).

Conclusions: This study provides evidence that the maintenance of a high or moderate CRF may slow the age-related decline in HRpeak in both men and women. The application of CRF-specific HRpeak prediction equations should be used to improve interpretation of HRpeak from exercise tests.

1Clinical Exercise Physiology Program, Human Performance Laboratory, Ball State University, Muncie, IN; and 2Department of Educational Psychology, Ball State University Muncie, IN

Address for correspondence: Cemal Ozemek, Ph.D., Mail Stop B179, 12631 East 17th Avenue, Room 8111, Aurora, CO 80045; E-mail: cemal.ozemek@ucdenver.edu.

Submitted for publication March 2015.

Accepted for publication July 2015.

Peak HR (HRpeak) is an import outcome measure of exercise testing as it, along with stroke volume, determines maximal cardiac output. It is well accepted that HRpeak declines with age, which is important to consider because failure to achieve an adequate HRpeak during a cardiopulmonary exercise test (CPET) is known to be an independent prognostic marker (18,19,36). The interpretation of HRpeak is typically referenced solely to age and includes a relatively high SD (12,13,29,34). The reported rate of decline has varied in different large cross-sectional data sets and in longitudinal examinations with small sample sizes (11,13,23,29). Whaley et al. (34) reported that resting HR, body weight or composition, physical activity status, and smoking history were all significantly related to HRpeak; however, the SE of the estimate of prediction was still relatively high (±11 bpm).

Another important outcome measure of exercise testing is the determination cardiorespiratory fitness (CRF or peak oxygen consumption (V˙O2peak)) (4,15). The relation between CRF and changes in HRpeak with age has not been thoroughly evaluated. An early cross-sectional study by Tanaka et al. (28) measured maximal exercise test responses in 84 endurance-trained athletes (age, 21–73 yr) with higher CRF and 72 nonobese sedentary subjects (age, 20–75 yr) with lower CRF and found no difference in the rate of decline with age. In contrast, a cross-sectional investigation by Trappe et al. (30) showed higher maximal HR values in a group (n = 9) of elite cross-country skier octogenarians (160 ± 5 bpm), with higher CRF compared with that in six age-matched sedentary individuals (146 ± 8 bpm), who had lower CRF. These equivocal findings may be partly due to the known limitations of cross-sectional study designs. Longitudinal studies on the rate of HRpeak decline with age between those with high and low CRF have also reported inconsistent findings.

Plowman et al. (24) found no significant difference in the rate of decline in HRpeak (roughly −0.5 bpm·yr−1) in women (23–65 yr) between CRF groups and across age. This study was limited by the short follow-up period (3.5–9 yr) and the small number of subjects within each age group (20–30 yr, n = 5; 31–40 yr, n = 8; 41–50 yr, n = 12; 51–60 yr, n = 6; 61–70 yr, n = 3). However, Rogers et al. (25) showed that master athletes (n = 15) with high CRF had no change in HRpeak over an 8-yr observation period. This was in contrast to a group of age-matched controls consisting of sedentary men (n = 14) with low CRF (age, 47–84 yr) who experienced an 8-bpm decline in HRpeak.

Some of the inconsistency in the studies of the age-related decline in HRpeak may be associated with how it is determined. The highest HR attained during an exercise test with a subjectively determined fatigue end point may not produce an accurate HRpeak. Assurance of a maximal or near-maximal effort should come from a CPET with the use of objective markers, most commonly the RER. Therefore, the primary aim of this study was to examine whether CRF contributes to the rate of decline in HRpeak in a large cohort that allows both cross-sectional and longitudinal comparisons with data collected from CPET.

Back to Top | Article Outline

METHODS

Subjects

Data were obtained from the Ball State University Adult Physical Fitness Program cohort and included 6033 CPET records from 3647 men and women (Fig. 1) between the years 1971 and 2014. Program participants were predominantly Caucasian, and most were sedentary at their initial test and are nonsmokers. All provided a written informed consent for cardiovascular risk screening and exercise testing and allowed for their deidentified data to be used in research investigations. This study was declared exempt from requiring institutional review board approval, as analyses were performed on deidentified retrospective data.

Inclusion criteria included the following: ≥18 yr of age, performing a CPET on a treadmill, and achieving an RER ≥ 1.0. Individuals taking medications that affect HR (β-blockers, antiarrhythmic, and/or digoxin) or with any medical history of heart disease (identified via health history questionnaire or physician report) were excluded. In addition, all cases that exhibited abnormal signs (i.e., angina, ST segment depression, significant dysrhythmia, or abnormal blood pressure response) leading to test termination before peak effort were excluded. Inclusion in the longitudinal analysis required participants to have completed ≥2 CPET with ≥1 yr between testing dates. Of the 3318 subjects included in the cross-sectional analysis, 643 subjects (418 men and 225 women) and 1872 tests met the inclusion criteria for longitudinal analysis. The average follow-up period (first to last CPET) was approximately 9.3 yr (ranging from 1 to 38.9 yr) with an average of 2.9 tests per person (range, 2–18 tests).

Back to Top | Article Outline

Laboratory Assessment

Participants were screened for medical contraindications to exercise using a health history questionnaire. The initial laboratory assessment served to identify traditional cardiovascular disease risk factors using standardized laboratory techniques that have been described in detail elsewhere (5,14,16,17,33,34).

All exercise tests were performed on a motorized treadmill. Standardized CPET protocols (i.e., Bruce (5), modified Bruce, BSU Bruce Ramp (17), Balke (3), modified Balke, and individualized walking/running protocols) were individually selected using subject characteristics to identify a protocol that would allow for the assessment of physiological responses to increasing intensities of exercise with a goal to achieve maximal effort in 8–12 min (1). The same protocol was performed by subjects in the longitudinal analyses unless there was a change in the individual’s characteristics that would suggest a different protocol would be more likely to achieve maximal effort in 8–12 min. Roughly half (48%) of the subjects included in the longitudinal analysis had at least one differing CPET protocol within their multiple CPET. CPET occurring before 1990 were commonly performed using an incremental-type protocol, whereas for those occurring after 1990, most of the tests were performed with the BSU-Bruce Ramp protocol. V˙O2peak was measured with metabolic measurements. Standardized procedures for metabolic cart calibration and V˙O2peak criteria were followed for all tests. All tests were supervised by experienced clinical exercise physiologists, and subjects were verbally encouraged to exercise to volitional fatigue. HR was continuously monitored via electrocardiogram, and the highest HR recorded during the CPET was used as the HRpeak.

Back to Top | Article Outline

Statistics

Descriptive statistics was used to characterize men and women within decade groupings (18–29 yr, 30–39 yr, 40–49 yr, 50–59 yr, and ≥60 yr) and within genders between decade groupings. All analyses were performed using SPSS version 22.0 (SPSS, Inc., Chicago, IL) except the linear mixed-models procedure, which was performed using R version 3.0.2 (R Foundation for Statistical Computing). This statistical approach is similar to that performed by Gellish et al. (12).

Back to Top | Article Outline

Cross-sectional analysis

Data from participants’ first CPET were used in the cross-sectional analysis. To test the effect of CRF on the decline in HRpeak with age, subjects were categorized into respective CRF categories (high, moderate, and low) on the basis of age and gender-specific normative CRF values established for this cohort. Individuals’ CRF ≤20th percentile in their gender age group were considered low fit, those in the ≥20th percentile of their age group, as highly fit, and individuals in the middle three quintiles were considered moderately fit. A general linear-model univariate analysis was performed to test the relation between CRF values and the decline in HRpeak with age. Linear regression analyses were then performed for the three CRF groups within the whole cohort and within gender, with HRpeak used as the dependent variable and age as the independent variable.

Back to Top | Article Outline

Longitudinal analysis

Descriptive statistics were used to characterize subjects’ resting and CPET responses from the first testing period between genders and within age groups. In addition, a multilevel, linear mixed-effects regression model was used to evaluate the change in subjects’ resting and CPET characteristics across age.

To assess the effects of the change in CRF over time on the decline in HRpeak with age, interactions between CRF group and age were tested, with HRpeak used as the dependent variable. Subject ID was treated as a random effect, which allowed testing values to be viewed almost as though they were collected from different fitness centers. Furthermore, the number of tests and time between tests for participants were not consistent throughout the cohort. The mixed-model procedure accounts for the variance associated with individuals, thereby using the full information provided by each study participant and yielding a natural weighting scheme in the estimate of overall fixed effects, such that those with more measurements contributed more information to these estimates. This statistical approach is similar to that performed by Gellish et al. (12). The effects of gender and CRF on HRpeak decline with age were tested by a general linear-model univariate analysis, which was performed to test for a three-way interaction among age, gender, and CRF.

Back to Top | Article Outline

RESULTS

Descriptive statistics and response to CPET for the cross-sectional analyses are presented in Table 1, and longitudinal analyses are presented in Table 2. Linear regressions between age and HRpeak for the cross-sectional and longitudinal analyses are presented in Figure 2 (A, B, C: cross-sectional; D, E, F: longitudinal).

In the examination of the effect of CRF on the decline in HRpeak with age, the cross-sectional analysis revealed a significant interaction (F = 1.338, P = 0.01) between CRF category and age. Subjects with a higher fitness (−0.60 bpm·yr−1) had a slower rate of decline in HRpeak compared with those with a lower fitness (moderate fitness, −0.78 bpm·yr−1; low fitness, −0.87 bpm·yr−1). The longitudinal analysis revealed results with a significant interaction (t-value = −6.37, P < 0.001) between CRF category and age (high fit, −0.61 bpm·yr−1; moderate fitness, −0.82 bpm·yr−1; and low fitness, −1.02 bpm·yr−1).

Analyses including gender, age, and CRF revealed a significant three-way interaction (F = 1.224, P = 0.01) in the cross-sectional analysis. Men had a faster rate of decline in HRpeak with age across CRF categories (high fitness, −0.62 bpm·yr−1; moderate fitness, −0.81 bpm·yr−1; low fitness, −0.95 bpm·yr−1) compared with women (high fitness, −0.57 bpm·yr−1; moderate fitness, −0.74 bpm·yr−1; and low fitness, −0.79 bpm·yr−1). In contrast, the longitudinal analysis did not reveal a significant three-way interaction among age, gender, and CRF (t-value = 1.28, P = 0.20).

Back to Top | Article Outline

DISCUSSION

It has been well established that exercise HRpeak declines across the life span. Although studies employing cross-sectional and/or longitudinal analytic designs have reported variation in the rate of decline with age, there is no debate regarding the presence of an inverse association. Numerous observational and experimental studies have assessed potential contributory factors to the decline in HRpeak (2,8,10,12,22–24,29,32,34), and many reports concluded that habitual physical activity and/or CRF did not serve as a modifier of the age-associated decline in HRpeak (10,12,23,24,29). However, Cooper et al. (8) reported that higher CRF was associated with attenuation of the age-associated decline in HRpeak in a large cohort of men tested at the Cooper Clinic. On the basis of the inconsistent findings and limitations found in previous studies, the aim of the present investigation was to reexamine the influence of CRF on the rate of decline in HRpeak with age in a large, apparently healthy, adult cohort of men and women. The present cohort represented a self-referred group that volunteered for a physical fitness assessment as a prelude to participation in a university-based exercise program.

The primary finding within the present study was a significant inverse association between CRF and the rate of decline in HRpeak across the age spectrum (cross-sectional design) and the life span (longitudinal design). Individuals within the present study who were classified as having high CRF had a slower rate of decline in HRpeak with age compared with moderate (second through fourth quintiles) and low (lowest quintile) fit individuals. This finding was observed in both the cross-sectional (n = 3318) and longitudinal (n = 643) analyses. The overall rates of decline in HRpeak within the longitudinal analyses were −0.84 and −0.78 bpm·yr−1 for men and women, respectively. These results indicate that the slope of −1.0 found within the commonly used 220-age formula (11) overestimates the decline in HRpeak in apparently healthy adults. The present study affirms this observation from several previous reports (8,12,23,24,29,34,37). However, our finding of an inverse association between CRF and HRpeak decline is in contrast to most previous reports. The rates of decline in HRpeak within the present study for high, moderate, and low fit individuals were −0.60, −0.78, and −0.87 bpm·yr−1 in the cross-sectional analysis, and −0.61, −0.82, and −1.02 bpm·yr−1 in the longitudinal analysis. These results indicate that the actual rate of decline for physically fit men and women is almost half a beat per year less than that predicted by the Fox et al. (11) equation.

Several previous studies have assessed the relation between CRF and/or physical activity and the age-related decline in HRpeak (8,10,12,23–25,29). Most relevant to the present study are those that reported an assessment of CRF and age-related decline in HRpeak (8,12,23–25). The results of the present study confirm those from Cooper et al. (8), who first reported the inverse association between CRF and decline in HRpeak in men tested at the Cooper Clinic. Although both studies included large cohorts of apparently healthy men, the present study included a large cohort of women in both the cross-sectional (n = 1469) and longitudinal (n = 225) analyses. Therefore, our results extend the finding of an inverse association between CRF and age-related HR decline to apparently healthy women. In addition to the inclusion of women in the present study, other differences between the two studies are worthy of mention. First, Cooper et al. (8) used a cross-sectional design, whereas the present study employed both cross-sectional and longitudinal analyses, which strengthens the findings. Second, the present study defined age- and gender-specific CRF categories using a measured V˙O2peak. Cooper et al. (8) defined CRF from peak treadmill time during a sign–symptom limited graded exercise test. Based on CRF data reported within the study of Cooper et al. (8), the mean predicted V˙O2peak (mL·kg−1·min−1) values for their male subjects were 14%, 5%, 3%, 1% lower, and 3% greater for the age decades of 20–29, 30–39, 40–49, 50–59, and ≥60 yr, respectively, when compared with similarly age men in the present study. Although the use of V˙O2peak as opposed to peak treadmill time does provide a more objective measure of CRF, we contend that the CRF across the age spectrum was reasonably similar between the two cohorts, and based on Figure 2 from the study of Cooper et al. (8), we conclude that the lowest fit group in each study had a slope for the decline most consistent with that of Fox et al. (11). Therefore, both reports using maximal exercise testing and large cohorts of apparently healthy men support the finding that CRF is inversely related to a preservation of HRpeak across the life span.

Our results are in contrast to other reports that assessed the association between a measure of CRF and the age-related decline in HRpeak (12,23–25). Gellish et al. (12) studied 132 men and women in a longitudinal design and reported no interaction between CRF and decline in HRpeak. Although they were the first to employ the longitudinal statistics used in the present study, their small sample size and lack of objective measures of CRF render their results less comparable with the present study’s findings that are based on objective measures of CRF. However, the authors characterized their sample as above average for CRF (>70th percentile) and their age coefficient (−0.67) lies between that of the high and moderate fit groups from the longitudinal analyses within the present study (−0.61 to −0.82 bpm·yr−1). Perhaps the combination of a 1) significantly larger sample, 2) more objective measure of CRF, and 3) wider range of CRF found within the present study allowed a more valid test of the relation between CRF and age-related decline in HRpeak. Nes et al. (23) reported that V˙O2peak did not significantly improve the age-adjusted prediction accuracy of HRpeak within a large (n = 3320) cross-sectional analysis of men and women. They defined CRF objectively and used a RER threshold (>1.05) to validate their exercise test end point. Similar to Gellish et al. (12), Nes et al. (23) characterized the sample in their analyses as “very fit.” By comparison, the cross-sectional cohort within the present study was slightly younger, significantly less fit, and less likely to be a smoker compared with the subjects in the study of Nes et al. (23). Whether these cohort differences can explain the disparate findings is unclear. It would seem as though age accounted for a similar amount of variance in the decline in HRpeak within both study analyses (r2 = 0.36). As Nes et al. (23) did not report age-adjusted coefficients for CRF, direct comparisons in the cross-sectional analyses cannot be made. Finally, using a longitudinal research design, Plowman et al. (24) followed a small group of women (n = 36) for between 3.5 and 9 yr and reported no differences in the decline in HRpeak across the CRF categories established within their analyses. The present study had a much larger female sample, and the CRF categories within the present study were adjusted for age, whereas this was not the case in the analyses of Plowman et al. (24).

There was a significant three-way interaction (gender–CRF–age) observed within the large cross-sectional analyses in the present study. Men had a steeper rate of decline per year across the CRF categories when compared with women. However, the longitudinal analyses within the present study did not confirm this gender–CRF interaction. Several previous studies with large samples of men and women (23,29) did not report differences between men and women in the rate of decline in HRpeak. However, Tanaka et al. (29) did not have CRF in their analyses, and Nes et al. (23) tested a CRF–age interaction within gender-specific analyses. It is unclear whether Nes et al. (23) assessed a model with CRF, gender, and age included. The current study was the first to test a three-way interaction between CRF, gender, and age in a large cohort of men and women, and the findings suggest that men may have a greater decline in HRpeak compared with women regardless of CRF. However, as previously mentioned, within gender, those with higher CRF experienced a slower age-related decline and this was affirmed within the longitudinal analyses. Future studies are needed to assess the gender differences in the cross-sectional analyses observed within the present study.

The findings from the present investigation have clear implications for estimating HR responses during exercise testing or when determining a target HR range for exercise prescriptions on the basis of a predicted HRpeak. Multiple large cohort studies (8,12,13,23,29,34) have shown that the 220 − age equation (11) underestimates HRpeak in older individuals and overestimates it in younger individuals. On the basis of the findings from the present study and those from Cooper et al. (8), CRF seems to influence the well-documented age-related decline in both adult men and women. Within both studies, the lowest fit group had a rate of decline most consistent with that defined by the 220 − age equation whereas higher fit individuals typically exceeded the predicted value. Again, the CRF association was observed in both our cross-sectional and longitudinal analyses. Clinicians should keep this in mind when using age-predicted HRpeak values in clinical practice. For example, using the high fit prediction equation for women from this study (HRpeak = 204 – 0.57 × age), a high fit 60-yr-old woman would be predicted to have an HRpeak of 170 bpm. The 220 − age formula would predict 160 bpm for the same women. On the basis of the findings from the present study and those of others that have shown a rate of decline less than 1.0 beat per year in adult men and women, the use of the 220 − age equation may only be appropriate when testing individuals that are thought to be fairly low fit. Finally, the differences in HRpeak observed across the CRF categories within the present study may have implications for the interpretation of the normal chronotropic response to maximal exercise.

A potential mechanism for the slower rate of decline in HRpeak with age in high fit individuals is the preservation of β-adrenergic receptor sensitivity. With age, β-adrenergic receptor sensitivity declines (7,9,31,35). Individuals with low CRF levels have been shown to have elevated levels of catecholamines at rest (6) and in response to exercise compared with endurance-trained individuals (20). The abnormally elevated level of circulating catecholamines has been found to further desensitize β-receptors (20). This in turn reduces the HRpeak one can achieve because the increase in HR beyond the maximal intrinsic rate is dependent on the sinoatrial node’s response to sympathetic stimulation. Regular exercise training has been shown to maintain a normal circulating level of catecholamines, preserving the sensitivity of the β-adrenergic receptors (21,27). This investigation can neither confirm nor reject the influence of catecholamines on β-adrenergic receptor desensitization. Therefore, future studies are needed to investigate the association of β-adrenergic receptor sensitivity and the age-related decline in HRpeak.

This study is not without limitations. The cohort found within this study was self-referred, apparently healthy, predominantly Caucasian. These characteristics therefore may limit the applicability of our findings to those with known cardiovascular disease and/or ethnic minorities. In addition, sufficient exercise effort in the current study was objectively determined by achieving an RER of ≥1.0. The standard criterion for a “maximal” effort is achieving an RER ≥ 1.1. However, there is a wide range of RER responses to CPET, even when maximal efforts are put forth. It is common for older individuals in particular to not attain RER values above 1.1 (26). As our cohort spans a wide age range, we believe that using an RER value of 1.0 is an appropriate criterion for determining peak effort. Despite these limitations, this study had many strengths. There was a large sample of subjects included in the cross-sectional and longitudinal studies that were used to observe the effects of CRF on the decline in HRpeak across a wide age range. The longitudinal analysis in the current study was the largest to investigate the effect of CRF on the age-related decline in HRpeak in both men and women. The implementation of the general linear model statistical method to the longitudinal data improved upon previous studies that evaluated the age-related decline within cross-sectional data. Furthermore, this study assessed a three-way interaction analysis (CRF–gender–age) and applied objective criteria to determine exercise efforts, unlike previous large cross-sectional and longitudinal studies that predicted CRF from exercise time. The use of objectively measured CRF to examine the effect of CRF on the age-related decline in HRpeak strengthens the outcomes of the current study compared with previous work.

Back to Top | Article Outline

CONCLUSIONS

In conclusion, this study provides evidence that the maintenance of a high or moderate CRF can slow the age-related decline in HRpeak in both men and women. Furthermore, in both the longitudinal and cross-sectional analyses, women tended to have less age-related decline in HRpeak as compared with men. The application of CRF-specific HRpeak prediction equations may improve the interpretation of the chronotropic response to maximal exercise.

The authors wish to thank the graduate assistants in the Clinical Exercise Physiology program at Ball State University for their assistance with data collection.

Funding for journal page charges was provided by the Ball State University Leroy “Bud” Getchell Graduate Student Professional Development Fund.

None of the authors have any conflicts of interest to declare.

The results of the present study do not constitute endorsement by the American College of Sports Medicine.

Back to Top | Article Outline

REFERENCES

1. ACSM’s Guidelines for Exercise Testing and Prescription. 9th ed: Lippincott Williams & Wilkins; 2013.
2. Astrand I. Aerobic work capacity in men and women with special reference to age. Acta Physiol Scand Suppl. 1960; 49 (169): 1–92.
3. Balke B, Ware RW. An experimental study of physical fitness of Air Force personnel. U S Armed Forces Med J. 1959; 10 (6): 675–88.
4. Blair SN, Kampert JB, Kohl HW 3rd, et al. Influences of cardiorespiratory fitness and other precursors on cardiovascular disease and all-cause mortality in men and women. JAMA. 1996; 276 (3): 205–10.
5. Bruce RA, Kusumi F, Hosmer D. Maximal oxygen intake and nomographic assessment of functional aerobic impairment in cardiovascular disease. Am Heart J. 1973; 85 (4): 546–62.
6. Chen M, Halter JB, Porte D Jr. Plasma catecholamines, dietary carbohydrate, and glucose intolerance: a comparison between young and old men. J Clin Endocrinol Metab. 1986; 62 (6): 1193–8.
7. Christou DD, Seals DR. Decreased maximal heart rate with aging is related to reduced {beta}-adrenergic responsiveness but is largely explained by a reduction in intrinsic heart rate. J Appl Physiol. 2008; 105 (1): 24–9.
8. Cooper KH, Purdy JG, White SR, Pollock ML, Linnerud AC. Age-fitness adjusted maximal heart rates. Med Sport. 1977; 10: 78–88.
9. Feldman RD, Limbird LE, Nadeau J, Robertson D, Wood AJ. Alterations in leukocyte beta-receptor affinity with aging. A potential explanation for altered beta-adrenergic sensitivity in the elderly. N Engl J Med. 1984; 310 (13): 815–9.
10. Fitzgerald MD, Tanaka H, Tran ZV, Seals DR. Age-related declines in maximal aerobic capacity in regularly exercising vs. sedentary women: a meta-analysis. J Appl Physiol (1985). 1997; 83 (1): 160–5.
11. Fox SM 3rd, Naughton JP, Haskell WL. Physical activity and the prevention of coronary heart disease. Ann Clin Res. 1971; 3 (6): 404–32.
12. Gellish RL, Goslin BR, Olson RE, McDonald A, Russi GD, Moudgil VK. Longitudinal modeling of the relationship between age and maximal heart rate. Med Sci Sports Exerc. 2007; 39 (5): 822–9.
13. Gulati M, Shaw LJ, Thisted RA, Black HR, Bairey Merz CN, Arnsdorf MF. Heart rate response to exercise stress testing in asymptomatic women: the st. James women take heart project. Circulation. 2010; 122 (2): 130–7.
14. Iribarren C, Jacobs DR Jr, Sidney S, et al. Serum total cholesterol and risk of hospitalization, and death from respiratory disease. Int J Epidemiol. 1997; 26 (6): 1191–202.
15. Kaminsky LA, Arena R, Beckie TM, et al. Circulation. 2013; 127 (5): 652–62.
16. Kaminsky LA, Whaley MH. Evaluation of within-day precision of serum cholesterol measured by a portable analyzer. Med Sci Sports Exerc. 1992; 24 (1): 134–8.
17. Kaminsky LA, Whaley MH. Evaluation of a new standardized ramp protocol: the BSU/Bruce Ramp protocol. J Cardiopulm Rehabil. 1998; 18 (6): 438–44.
18. Khan MN, Pothier CE, Lauer MS. Chronotropic incompetence as a predictor of death among patients with normal electrograms taking beta blockers (metoprolol or atenolol). Am J Cardiol. 2005; 96 (9): 1328–33.
19. Lauer MS, Francis GS, Okin PM, Pashkow FJ, Snader CE, Marwick TH. Impaired chronotropic response to exercise stress testing as a predictor of mortality. JAMA. 1999; 281 (6): 524–9.
20. Lehmann M, Schmid P, Keul J. Age- and exercise-related sympathetic activity in untrained volunteers, trained athletes and patients with impaired left-ventricular contractility. Eur Heart J. 1984; 5 (E Suppl): – .
21. Leosco D, Rengo G, Iaccarino G, et al. Exercise training and beta-blocker treatment ameliorate age-dependent impairment of beta-adrenergic receptor signaling and enhance cardiac responsiveness to adrenergic stimulation. Am J Physiol Heart Circ Physiol. 2007; 293 (3): H1596–603.
22. Lester M, Sheffield LT, Trammell P, Reeves TJ. The effect of age and athletic training on the maximal heart rate during muscular exercise. Am Heart J. 1968; 76 (3): 370–6.
23. Nes BM, Janszky I, Wisloff U, Stoylen A, Karlsen T. Age-predicted maximal heart rate in healthy subjects: The HUNT fitness study. Scand J Med Sci Sports. 2013; 23 (6): 697–704.
24. Plowman SA, Drinkwater BL, Horvath SM. Age and aerobic power in women: a longitudinal study. J Gerontol. 1979; 34 (4): 512–20.
25. Rogers MA, Hagberg JM, Martin WH 3rd, Ehsani AA, Holloszy JO. Decline in VO2max with aging in master athletes and sedentary men. J Appl Physiol (1985). 1990; 68 (5): 2195–9.
26. Sidney KH, Shephard RJ. Maximum and submaximum exercise tests in men and women in the seventh, eighth, and ninth decades of life. J Appl Physiol Respir Environ Exerc Physiol. 1977; 43 (2): 280–7.
27. Spina RJ, Turner MJ, Ehsani AA. Beta-adrenergic-mediated improvement in left ventricular function by exercise training in older men. Am J Physiol. 1998; 274 (2 Pt 2): H397–404.
28. Tanaka H, Desouza CA, Jones PP, Stevenson ET, Davy KP, Seals DR. Greater rate of decline in maximal aerobic capacity with age in physically active vs. sedentary healthy women. J Appl Physiol (1985). 1997; 83 (6): 1947–53.
29. Tanaka H, Monahan KD, Seals DR. Age-predicted maximal heart rate revisited. J Am Coll Cardiol. 2001; 37 (1): 153–6.
30. Trappe S, Hayes E, Galpin A, et al. New records in aerobic power among octogenarian lifelong endurance athletes. J Appl Physiol (1985). 2013; 114 (1): 3–10.
31. Vestal RE, Wood AJ, Shand DG. Reduced beta-adrenoceptor sensitivity in the elderly. Clin Pharmacol Ther. 1979; 26 (2): 181–6.
32. Wergel-Kolmert U, Wisen A, Wohlfart B. Repeatability of measurements of oxygen consumption, heart rate and Borg’s scale in men during ergometer cycling. Clin Physiol Funct Imaging. 2002; 22 (4): 261–5.
33. Whaley MH, Kaminsky LA, Dwyer GB, Getchell LH, Norton JA. Predictors of over- and underachievement of age-predicted maximal heart rate. Med Sci Sports Exerc. 1992; 24 (10): 1173–9.
34. Whaley MH, Kaminsky LA, Dwyer GB, Getchell LH. Failure of predicted VO2peak to discriminate physical fitness in epidemiological studies. Med Sci Sports Exerc. 1995; 27 (1): 85–91.
35. White M, Leenen FH. Aging and cardiovascular responsiveness to beta-agonist in humans: role of changes in beta-receptor responses versus baroreflex activity. Clin Pharmacol Ther. 1994; 56 (5): 543–53.
36. Wilkoff BL, Miller RE. Exercise testing for chronotropic assessment. Cardiol Clin. 1992; 10 (4): 705–17.
37. Zhu N, Suarez-Lopez JR, Sidney S, et al. Longitudinal examination of age-predicted symptom-limited exercise maximum HR. Med Sci Sports Exerc. 2010; 42 (8): 1519–27.
Keywords:

EXERCISE TESTING; AGING; CROSS-SECTIONAL; LONGITUDINAL

© 2016 American College of Sports Medicine