Clinical utility of maximal heart rate

In clinical practice, maximal heart rate is used to guide exercise prescription and exercise training intensity. In chronic cardiovascular conditions, such as inherited arrhythmic syndromes, maximal heart rate derived through exercise testing is an essential method of determining the degree of β-blockade and other therapy prescribed to limit exertional heart rate and prevent exertional ventricular arrhythmias occurring. ^{Reference El Assaad, Heilbronner, Zahka, Hammond, Patel and Aziz15} In children, doses of β- blocker therapy are up-titrated following exercise testing, with the maximal heart rate achieved during testing used to identify if doses remain at a sufficient level as the child grows and as their heart rate response to exercise changes. Accurate measurement of maximal heart rate is an important component of testing to ensure adequate pharmacological control and reduction of risk.

Measurement of maximal heart rate

Direct measurement of maximal heart rate in children is achieved through maximal exercise testing; however, in children taking therapy that limits chronotropic response to exercise, indirect methods of predicting maximal heart rate are utilised, where the maximal heart rate achieved during testing is expressed as a percentage of the expected predicted maximal heart rate value. In most clinical settings, the same prediction equation, Fox (220-age), ^{Reference Fox, Naughton and Haskell5} is used during exercise testing for adults and children; however, methodological concerns have been raised about this equation in both adults and children. ^{Reference Robergs and Landwehr16} In line with the findings of previous childhood studies, ^{Reference Nikolaidis1,Reference Heinzmann-Filho, Zanatta and Vendrusculo6–Reference Papadopoulou, Papadopoulou and Alipasali8,Reference Mahon, Marjerrison, Lee, Woodruff and Hanna10–Reference Colantonio and Kiss13} the results of the current study demonstrated that the Fox equation significantly overpredicts maximal heart rate in both males and females, suggesting that this is not an accurate method for assessing maximal heart rate in children.

Sex and age differences in maximal heart rate

Maximal heart rate regression equations, including Fox, ^{Reference Fox, Naughton and Haskell5} are largely devised for use in adults, based on the assumption that maximal heart rate declines with age; however, children’s exertional heart rates do not decline with age at the same rate as in adults. ^{Reference Washington, Bricker and Alpert17} Furthermore, children’s exertional heart rates are generally higher than adults, and there are important sex differences. Girls demonstrate higher heart rates at submaximal exertion; however, no differences in maximal heart rate between boys and girls have been reported. ^{Reference Winsley18} Post-puberty and once maturation is achieved, maximal heart rate is thought to decline by approximately 1 beat per minute per year in males and females, in line with adult heart rate decline. Sex-specific prediction equations have been proposed, but these are also intended for use in adult populations. ^{Reference Whyte, George, Shave, Middleton and Nevill19–Reference Edvardsen, Hansen, Holme, Dyrstad and Anderssen23}

Our data show that maximal childhood heart rates during exercise over 200 beats per minute are common, with higher values possible, albeit with large inter-individual variation. We also saw differences in the relationship between age, heart rate response, and sex.

Males’ maximal heart rate increased with age, resulting in a small to moderate, negative significant correlation with all predicted maximal heart rate values. Similarly, a statistically significant mean difference for most equations was apparent when compared with measured mean maximal heart rate, suggesting those equations overpredict. In contrast, in females, maximal heart rate appeared to decline slightly with age, resulting in a small positive correlation with maximal heart rate and some prediction equations. Similarly, statistically significant mean differences were seen with most equations and overprediction of maximal heart rate. The data suggest that paediatric sex-specific maximal heart rate equations are required to assess exertional heart rate response during childhood.

Clinical implications

When heart rate prediction equations are applied for exercise prescription purposes, errors in estimating maximal heart rate can result in inaccurate prescription of cardiorespiratory effort during exercise. ^{Reference Heinzmann-Filho, Zanatta and Vendrusculo6} Further, outside the exercise prescription setting, the implications of overprediction of maximal heart rate in children’s clinical cardiology practice during exercise testing may result in a lower percent predicted maximal heart rate value being calculated, which is then utilised to determine whether dose up-titration is required. Studies investigating the applicability of the Fox equation in child athletes found that Fox overpredicts maximal heart rate. ^{Reference Nikolaidis1,Reference Machado and Denadai7,Reference Verschuren, Maltais and Takken11,Reference Cicone, Holmes, Fedewa, MacDonald and Esco24,Reference A., Aniuska, dM, dC, Aline and Paulo25} The findings from the present study also demonstrated that this is also the case for both male and female non-athletic children. If using the Fox equation to predict maximal heart rate in a 5-year-old child, the predicted maximal heart rate would be 215 beats per minute. If the maximal heart rate achieved during testing was 150 beats per minute, this would be reflected as 70% of the predicted maximal heart rate achieved. In the current study, Fairbarn and Whyte predicted most closely to measured maximal heart rate. If using either of these equations, for example, the predicted maximal heart rate would be 197.85 beats per minute (76%) or 199.25 beats per minute (75%), respectively, resulting in a 5–6% difference compared with Fox. Depending on the cut-off for β-blocker adjustment, which is also subject to variation between clinicians, this difference could influence the dose of β-blocker being deemed as sufficient, suggesting a dose increase is not required. Whereas if using an equation that predicts more closely to measured maximal heart rate, a dose increase would be advised.

Other studies have focused attention on the equation of Tanaka et al., ^{Reference Tanaka, Monahan and Seals26} with some suggesting this as a more accurate equation to use in children and adolescents ^{Reference Mahon, Marjerrison, Lee, Woodruff and Hanna10}, whereas others have demonstrated an underestimation of predicted maximal heart rate. ^{Reference Nikolaidis1,Reference Papadopoulou, Papadopoulou and Alipasali8,Reference Nikolaidis, Padulo, Chtourou, Torres-Luque, Afonso and Heller12} In the current study, the Tanaka equation also significantly overpredicted in both males and females. Alternative age regression prediction equations evaluated for validity and applicability in children and adolescents have all been found to demonstrate a low correlation with measured maximal heart rate. ^{Reference Gelbart, Ziv-Baran, Williams, Yarom and Dubnov-Raz9,Reference Cicone, Holmes, Fedewa, MacDonald and Esco24,Reference Cicone, Sinelnikov and Esco27}

A novel finding in our study is the demonstration that sex-specific and discordant differences exist between mean measured and predicted maximal heart rate. In females, most equations appeared to overpredict, but two equations underpredicted: Fairbarn (devised specifically for females) and Heinzmann-Filho (devised specifically for obese adolescents). The equation of Whyte, ^{Reference Whyte, George, Shave, Middleton and Nevill19} an equation established for male athletes, was found to predict most closely in females; whereas in males, the equation of Fairbarn, originally devised for use in females, was found to most closely predict maximal heart rate. Together, these findings demonstrate that existing prediction equations are not suitable for use in the paediatric population and highlight the importance of developing paediatric-specific and sex-specific maximal heart rate prediction equations as a clinical priority.

Limitations

This study is limited by a relatively small sample size, although the numbers are similar to other studies testing these equations in children and adolescents. ^{Reference Nikolaidis1,Reference Heinzmann-Filho, Zanatta and Vendrusculo6,Reference Machado and Denadai7,Reference Mahon, Marjerrison, Lee, Woodruff and Hanna10,Reference Nikolaidis, Padulo, Chtourou, Torres-Luque, Afonso and Heller12,Reference Colantonio and Kiss13,Reference Cicone, Sinelnikov and Esco27,Reference Nikolaidis28} Similarly, many of the prediction equations were derived from relatively small sample sizes.

There was variation in the length of time children remained on the exercise treadmill, resulting in shorter testing times and lower maximal heart rate, possibly representing a submaximal test, and thus potentially impacting the range in maximal heart rates achieved; however, the included exercise tests were representative of the wider population. The likelihood that children attain a true maximal heart rate during exercise testing has also been questioned, ^{Reference Robergs and Landwehr16} suggesting that other forms of testing may produce higher maximal heart rates.

Finally, pubertal stage, which may account for the decline in heart rate seen in some of the females, was not evaluated in this study and requires additional research. ^{Reference Cicone, Holmes, Fedewa, MacDonald and Esco24} In this context, previous studies have suggested that other anthropometric variables may impact maximal heart rate requiring additional research. ^{Reference Cicone, Holmes, Fedewa, MacDonald and Esco24,Reference Carli, Moraes Junior, Menezes-Junior, Tadiotto, Mota and Leite29}

Conclusion

This study shows that adult age regression equations do not accurately predict childhood maximal heart rate. In the absence of direct measurement, alternative methods of predicting maximal heart rate should be defined to ensure appropriate estimation when prescribing pharmacological therapy and during subsequent exercise prescription particularly for children with inherited arrhythmic syndromes.