The results of this study indicated that higher RHR was related to lower cardiorespiratory fitness, independent of obesity and other confounders. Moreover, the inclusion of cardiorespiratory fitness as a confounder in logistic regression made the associations of the outcomes with RHR non-significant. Previous studies have reported that the relationship between cardiorespiratory fitness and lipid variables/blood pressure in adolescents is mediated by body fatness, whereas the observed relationships with fatness are independent of cardiorespiratory fitness [20
]. Our findings indicate an inverse effect of these confounders on the relationship between RHR and the metabolic outcomes (independent of fatness and strongly dependent on cardiorespiratory fitness).
The close inverse relationship between cardiorespiratory fitness and RHR has been demonstrated in previous reports by other authors [10
]. The recognized effect of cardiorespiratory fitness in autonomic nervous system activity and subsequent adaptations in neurohumoral control (decrease in circulating levels of catecholamines and changes in number or affinity of receptors) [21
] seems to be independent of body composition [22
] and could offer support to our results.
A previous study [22
] found that parasympathetic indexes of obese adults engaged in ≥2 hours per week of physical exercises were higher than those observed in sedentary adults of normal weight. Moreover, this protective effect has been identified in children. Gutin et al. [23
] identified an improvement in parasympathetic activity in obese children submitted to 8 months of a physical training protocol, which decreased after subsequent detraining (changes in parasympathetic activity were not related to modifications in body fatness). In our study, cardiorespiratory fitness was negatively related to RHR (sport practice only in the univariate model) and, therefore, as previously observed in other cardiovascular and metabolic outcomes [1
], physical activity practice during early life could be useful in the prevention of excessive weight gain [20
], promotion of lower RHR and hence the prevention of cardiovascular diseases in adulthood.
Additionally, high RHR was also associated with sleep pattern. Recently, Gallicchio and Kalesan [24
] in a systematic review/meta-analysis identified that people with both short and longer periods of sleep are at an increased risk of all-cause mortality. However, the actual pathway by which sleep is linked to cardiovascular complications [25
] is not clear, although it is plausible to believe that a pathway exists. Adolescents are prone to perform more activities at night (TV viewing and computer usage) than children and thus they are more exposed to shorter periods of sleep. Short sleep may act as an acute and chronic stressor and, therefore, affect the sympathetic activity of the organism and lead to an increase in RHR [26
]. Moreover, the concentration of pro-inflammatory agents (interleukine-6, tumor necrosis factor – alpha and C-reactive protein) is increased in people with short sleep periods [26
]. Our findings highlight the fact that health professionals must target the promotion of adequate sleep patterns among pediatric populations, because this harmful relationship between sleep pattern and a higher RHR seems significant from an early age.
RHR has significant potential for screening increased fasting glucose values. In agreement with this, Dubose et al. [27
] recently identified that RHR can be used, together with other variables, to screen American adolescents with glucose impairment. Researches have indicated that insulin resistance has an important relationship with sympathetic activation [28
], which significantly affects the RHR.
Similarly, dysfunction in lipid metabolism was also related to a high RHR. A previous study [11
] found a positive and significant relationship between RHR, triglycerides and TC among obese children and adolescents. On the other hand, the same authors point out that the causality/pathway by which a high RHR is linked to lipid dysfunction cannot be clearly determined. It is plausible to believe that insulin resistance could also be relevant in this process [28
]. In fact, insulin resistance affects the process of energy production, leading to an increased use of lipids as fuel and a higher production of reactive oxygen species in the brain (by the activation of the nicotine adenine dinucleotide hydrogen phosphatase oxidase), which increases the oxidative stress in the rostral ventrolateral medulla, the region that determines the basal sympathetic activity [29
]. Apparently, this inflammatory process occurs irrespective of the presence of obesity and ratifies the potential of RHR for screening adolescents at an increased cardiovascular risk.
] pointed out that among adults there are no doubts that an RHR ≥80 to 85 beats per minutes implies an increased risk for health. In pediatric populations this RHR range seems not to be true, because the cutoffs were different according to outcomes analyzed (ranging from 81.5 to 89). Moreover, previous studies (and also our findings) have reported a significant RHR variation according to age groups [9
]. Therefore, future cutoff tables should be developed in longitudinal observations and take into account adjustments for gender and age.
Our study has strengths, such as the sample size calculation and random process for selecting the schools/classes. However, the limitations must be recognized too. Initially, the absence of dietary habits related to RHR (e.g. cola intake, coffee, energy drinks) constitutes a significant weakness in our study and a target for further investigations. Our sample has a wide age range and the peak height velocity has limitations when applied to some age groups within this range. On the other hand the use of other methods to estimate pubertal stage involves ethical and logistical complications. The absence of measures related to adipokines and insulin resistance must be recognized and should be the focus of future investigations.
Finally, the low magnitude of the correlation coefficient found [32
] should be taken into account in further inferences, because it denotes the action of other variables in the relationship between RHR and the analyzed outcomes. Thus, mediated effect could be controlled by the simultaneous use of the RHR together with other variables (e.g. general obesity, abdominal obesity, low cardiorespiratory fitness) to screen adolescents at an increased metabolic risk in further studies [27