Resting heart rate but not heart rate variability is associated with the normal‐weight obesity phenotype

To determine differences in resting heart rate variability (HRV) and heart rate (HR) between young adults with normal‐weight obesity (NWO) and normal‐weight lean (NWL).

Normal-weight obesity (NWO) refers to individuals that present a normal weight, according to body mass index (BMI) values, but have an excess of adiposity (Wijayatunga & Dhurandhar, 2021).The worldwide prevalence for NWO ranges from 4.5% to 22% (Wijayatunga & Dhurandhar, 2021).In contrast, normal-weight lean (NWL) individuals show lower adiposity levels, more lean mass, and a better cardiovascular disease (CVD) risk profile than NWO (Wijayatunga & Dhurandhar, 2021).NWO is associated with an increased risk of metabolic disturbances, cardiovascular issues, and other health complications typically attributed to excess body fat, despite having a normal body weight according to traditional standards (Wijayatunga & Dhurandhar, 2021).Moreover, NWO is associated with a higher risk of CVD mortality compared with NWL (Romero-Corral et al., 2010;Wijayatunga & Dhurandhar, 2021).
In the clinical context, it is interesting to have and use noninvasive tools/biomarkers, which are relatively easy and fast to measure and cheap, to identify young individuals who are apparently healthy but at risk of developing CVD.Resting heart rate variability (HRV), which is the variation in time between consecutive heart beats, and indicates the modulation of the autonomic nervous system on the sinoatrial node activity (Task Force, 1996), is a widely employed noninvasive and health-related biomarker.Lower values on resting vagalrelated HRV parameters (hereinafter called "HRV") can predict the first cardiovascular adverse event in apparently healthy adults (Hillebrand et al., 2013).Furthermore, HRV is inversely associated with traditional CVD risk factors and obesity, determined by BMI cutoff points, in young adults (Plaza-Florido, Sacha, & Alcantara, 2022;Strüven et al., 2021).Likewise, closely related to HRV is resting mean heart rate (HR; Plaza-Florido, Migueles, Sacha, & Ortega, 2019), which indicates the number of heart beats per minute and is a well-known CVD risk factor (Aune et al., 2017;Tadic et al., 2018).An elevated HR is related to a higher risk of CVD and all-cause mortality (Aune et al., 2017;Tadic et al., 2018).To better understand, whether both HRV and HR are associated with the NWO phenotype is of clinical and public health interest.As of our current understanding, the potential similarity of HRV and HR in individuals exhibiting the NWO phenotype versus those with the NWL phenotype remains unexplored.The answer to this question interests clinicians because it might provide noninvasive and nonexpensive biomarkers to detect young adults with healthy body weight but presenting an adverse CVD risk profile.Thus, we explored differences in HRV and HR between young adults with NWO and NWL.We hypothesize that individuals with NWO will present lower HRV and higher HR compared with NWL.

| Participants and study design
Sixty-five adults (50 women, 77%) aged 18-25 years old (students from the University of Granada) with normal weight (BMI 18.5 to 24.9 kg/m 2 ) and a sedentary lifestyle (< 20 min moderate-vigorous physical activity <3 days/ week) from the project (Clinicaltrial.gov.ID: NCT02365129, (Sanchez-Delgado et al., 2015)) were included in this ancillary cross-sectional study.The Research Ethic Committee at the University of Granada (no.924) approved the study procedures, which followed the Declaration of Helsinki.All participants provided written informed consent.Whole-body fat percentage (BF%), lean mass, and visceral adipose tissue were determined using a whole-body dual energy x-ray absorptiometry scanner (Discovery Wi, Hologic Inc., Bedford, MA, USA).Body weight and height were assessed using a scale and a stadiometer (SECA model 799, Hamburg, Germany), and BMI was calculated.There are different cutoff points to define NWO using BF%, but there is no consensus yet about the "best" cutoff point (Wijayatunga & Dhurandhar, 2021).We used cutoff points for BF% derived from a large population study that included 6171 adults (BF% >33.3 for women and >23.1 for men, respectively) (Romero-Corral et al., 2010).Traditional CVD risk factors (total cholesterol, high-density and low-density lipoprotein-cholesterol, triglycerides, glucose, insulin levels, and the homeostatic model of insulin resistance [HOMA] index) were determined in blood after overnight fast (12 h), as described previously (Plaza-Florido et al., 2021).

| Resting HRV and HR
The HRV assessment, data processing and derivation of HRV parameters were extensively described elsewhere (Plaza-Florido et al., 2021).Briefly, HRV was assessed in the morning (8-9 AM), in the supine position, on a bed.The Polar RS800CX (Polar Electro, Kempele, Finland) and the Kubios HRV software (v.3.0.0 [free version]) were used to assess and analyze the HRV signal (Tarvainen et al., 2014), respectively.From the 15-min HRV recording, we selected the "best" 5-min segment and used the medium Kubios filter following previous recommendations (Alcantara et al., 2020).The sampling frequency was established at 1000 Hz and the alpha at 500.We computed HRV parameters in (i) time domain: the squared root of the mean of the sum of the squares of successive normal R-R interval differences (RMSSD), the standard deviation of all normal R-R intervals (SDNN), and the percentage number of pairs of adjacent normal R-R intervals differing by more than 50 ms in the entire recording (pNN50); and, in (ii) frequency domain: the power in the high frequency (HF) band 0.15-0.4Hz in absolute units (ms 2 ).In addition, we measured HR, which represents the number of heart beats per minute (bpm).

| Statistical analyses
The SPSS software v. 22.0 (IBM Corporation, Chicago, IL, USA) was used to perform statistical analyses.Shapiro-Wilk test and the visual inspection of histograms were used to determine the variables that showed a skewed distribution, and the logarithmic transformation was performed for these variables.Analysis of variance (ANOVA) was conducted to examine the differences in body composition, traditional CVD risk factors, HRV, and HR between NWO and NWL groups.There was no sex Â group interaction, therefore, the analyses were conducted in men and women together.Statistical significance was set at p < .05.

| RESULTS
Table 1 shows an adverse CVD risk profile (i.e., higher triglycerides, visceral adiposity, and insulin resistance) in NWO compared with NWL individuals (all p < .05).No differences in HRV parameters were observed between the NWO and NWL groups (Figure 1A-D).The NWO group presented, however, higher levels of HR (mean difference: 5 bpm) than the NWL group (95% CI, 0 to 10 bpm, Figure 1E).The results did not change when we repeated the analyses in men and women separately (women and men: mean difference: 5.8 and 4.6 bpm, respectively).

| DISCUSSION
This study showed that young adults with NWO presented higher HR levels than those with NWL.However, we did not detect differences in HRV between the NWO and NWL phenotypes.These findings reinforce the utility of HR as a potential biomarker in the "apparently healthy" NWO phenotype.Similar to previous publications (Wijayatunga & Dhurandhar, 2021), we showed that young adults with NWO presented an adverse CVD risk profile compared with NWL (i.e., higher triglycerides, visceral adiposity, and insulin resistance), whereas HR was higher in the former.This observation is interesting because HR, rather than HRV, has been better associated with CVD risk factors such as cardiorespiratory fitness in young adults (Grant et al., 2013;Plaza-Florido, Amaro-Gahete, Acosta, Sacha & Alcantara, 2022).In this context, HR is a noninvasive cardiac autonomic indicator requiring less technical demands and time to be processed, analyzed, and interpreted than HRV, which can be expressed using tens of parameters in different domains (Task Force, 1996).This is relevant for clinicians, particularly if we consider that HR is a well-known risk factor for developing CVD and mortality (Hozawa et al., 2004;Tadic et al., 2018).Our study identified a mean HR difference of 5 bpm (confidence intervals reached 10 bpm) between NWO and NWL groups, which could have clinical relevance.In fact, during resting conditions, an increase of 5 bpm was related to a higher (+17%) CVD mortality risk (Hozawa et al., 2004), whereas an increase of 10 bpm was associated with a higher risk (≈20%) of cardiac dying for cardiovascular complications (Benetos et al., 1999).Moreover, a meta-analysis highlighted that an increase of 10 bpm in resting HR was related to a higher risk of cardiovascular morbidity and higher all-cause mortality (around 15% and 17%, respectively) (Aune et al., 2017).This issue has been extensively discussed elsewhere (Tadic et al., 2018).Recently, a pilot study showed that a 4-week exercise intervention reduced resting HR in women with NWO, suggesting an improved CVD risk profile (Hu et al., 2022).Interestingly, a previous study in young women reported an inverse association between HRV and adiposity (Triggiani et al., 2019), whereas in our study, there were no differences in HRV between NWO and NWL groups despite the former presenting higher adiposity levels.At the same time, one study in children showed that the association between HRV and  ) shows the resting mean heart rate (HR) in beats per minute (bpm).All HRV parameters were transformed using the natural logarithm (ln).* represents p value <.05 derived from analysis of covariance (ANCOVA) to examine between groups (i.e., NWO vs. NWL) adjusted mean differences while considering participants' age and sex into account as potential confounders (i.e., ANCOVA models adjusted for age and sex).N = 41 and 24 for NWO and NWL, respectively.adiposity was explained in part by HR (Plaza-Florido et al., 2019;Plaza-Florido, Migueles, Sacha & Ortega, 2019).Thus, more studies are warranted on this research topic to confirm or contrast our findings.
Our study presents some limitations that should be acknowledged.First, the cross-sectional design does not allow us to infer causality from the associations reported.Second, we used an HR monitor to assess HRV instead of an electrocardiograph, which is the reference tool.Third, we did not consider breathing frequency during the HRV assessment, which could have influenced our findings.Fourth, the BF% cutoff points provided by Romero-Corral et al. (2010) were calculated in a more heterogeneous sample of young and middle-aged adults, whereas in our study, there was a narrower range of young adults; this issue could introduce some bias.Despite these limitations, we followed a solid and well-standardized methodology to assess and analyze HRV based on previous recommendations (Alcantara et al., 2020).Furthermore, to our knowledge, we reported the null association between HRV and NWO phenotype for the first time, and that HR was higher in young adults with NWO compared with NWL.In summary, our findings suggest that despite being normalweight (as categorized by BMI), NWO phenotype is associated with an adverse CVD risk profile even in "apparently healthy" young adults.

| CONCLUSION
Young adults with NWO presented higher HR than individuals with NWL at resting conditions, whereas there were no differences for HRV.These findings suggest that HR should be considered a potential noninvasive risk factor for CVD in young adults with NWO.Future studies using larger sample sizes and longitudinal designs should contrast or confirm our findings and perform analyses in men and women separately.

F
I G U R E 1 Column plots for each heart rate variability (HRV) parameter separated by individuals with normal-weight obesity (NWO; orange circles) and with normal-weight lean (NWL; blue circles).Results are presented as mean and standard deviation and all data points (i.e., individual values).Panel (A) shows the squared root of the mean of the sum of the squares of successive normal R-R interval differences (RMSSD).Panel (B) shows the standard deviation of all normal R-R intervals (SDNN).Panel (C) shows the percentage number of pairs of adjacent normal R-R intervals differing by more than 50 ms in the entire recording (pNN50).Panel (D) shows the power in the high frequency (HF) band.Panel (E Mean values and differences in age and traditional cardiovascular disease risk factors between young adults with normalweight obesity and normal-weight lean. T A B L E 1Note: Values are expressed as mean (95% confidence interval).Lean mass, high-density lipoprotein, and triglycerides are presented in logarithmic (ln) units.Abbreviations: BMI, body mass index; HOMA, homeostasis model assessment; NWL, normal-weight lean; NWO, normal-weight obesity.*p < .05.