Autonomic dysfunction in individuals with metabolically healthy obesity: a cross-sectional study

Abstract

The identification of autonomic alterations in metabolically healthy obesity could be an additional parameter for elucidating if this phenotype of obesity only represents a transition stage to the development of metabolically unhealthy obesity. So, this study aimed to assess the presence of autonomic dysfunction in individuals with metabolically healthy obesity. We conducted a cross-sectional study that randomly recruited individuals (≥18 years old), being excluded those on sick leave, pregnant or lactating women. According to the body mass index and International Diabetes Federation criteria for metabolic syndrome, the individuals, stratified as metabolically healthy eutrophic; metabolically healthy obesity; or metabolically unhealthy obesity were included in the analysis. Heart rate variability (HRV), electrocardiography, orthostatic test, and orthostatic hypotension test were used to evaluate autonomic function. Our findings evidenced that metabolically healthy obesity individuals exhibited alterations in HRV, in orthostatic hypotension test and in electrocardiographic parameters for ventricular repolarization, compared to metabolically healthy eutrophic individuals. Therefore, we concluded that metabolically healthy obesity is associated with the occurrence of autonomic dysfunction, which suggests that it can be, in fact, a transition stage to the development of metabolically unhealthy obesity.

Key words
Autonomic Dysfunction; Autonomic Nervous System; Heart Rate Variability; Metabolically Benign Obesity; Metabolically Healthy Obesity; Orthostatic Hypertension

INTRODUCTION

Obesity, a global public health concern, is recognized as a worldwide epidemic since 1997 given its increasing prevalence (Chooi et al. 2019, Haththotuwa et al. 2020, Boutari & Mantzoros 2022). According to the World Health Organization’s (WHO) report on “Accelerating action to stop obesity”, over 1 billion people worldwide are affected by obesity, comprising 39 million children, 340 million adolescents, and 650 million adults (WHO 2022).

Despite the extensive literature emphasizing obesity as an independent cardiovascular disease risk factor (Akil & Ahmad 2011, Ebong et al. 2014, Barroso et al. 2017, Csige et al. 2018, Manrique-Acevedo et al. 2020), in 1980’s decade some studies (Keyes 1973, Andres 1980, Sims 1982) described a “benign” obesity phenotype called metabolically healthy obesity (MHO). MHO, also called as metabolically benign obesity, represents a phenotype of obesity segregated from the metabolic syndrome criteria, which make emerge the hypothesis/common sense that MHO could be associated with lower cardiovascular risk in comparison to those individuals with obesity that present metabolic abnormalities (Samocha-Bonet et al. 2014). However, there remains a lack of consensus in the literature regarding weather MHO truly represents a benign obesity profile associated with lower cardiovascular risk and mortality or if it simply signifies a transitional stage to development of obesity-related metabolic abnormalities (Mongraw-Chaffin et al. 2018).

Moreover, some studies attribute the high prevalence of cardiovascular diseases in individuals with obesity not only to the presence of metabolic disorders but also to autonomic (Akhter 2011, Yadav et al. 2017). In fact, it has been described that autonomic dysfunction can precede the development of type 2 diabetes, and also predict metabolic syndrome development (Yu & Lee 2021). This hypothesis is supported by the positive association between the increase in the waist-to-hip ratio — an indicator of visceral adiposity — and the reduction in parasympathetic activity with an increase in sympathetic activity, representing what is termed as sympathetic-vagal imbalance, which elevates cardiovascular risk in individuals with obesity (Yadav et al. 2017). Therefore, due to sympathetic overactivation and sympathetic-vagal imbalance observed in obesity, the early identification of these alterations could be an advice to enhance the treatment and monitoring of these individuals, aiming to reduce the development of cardiovascular diseases and other metabolic disorders (Canale et al. 2013, Guarino et al. 2017).

Although some studies revealed reduced parasympathetic activity and sympathovagal imbalance even in the absence of metabolic disorders, using heart rate variability (Chintala et al. 2015, Rastović et al. 2016), other autonomic function parameters, such as hemodynamic responses and electrocardiographic alterations, remains uncertain in MHO, underscoring a critical gap that justifies the novelty of our investigation into comprehensive autonomic dysfunction in MHO.

Likewise, the identification of autonomic alterations in individuals with obesity without metabolic disorders could be an additional parameter for elucidating the transient nature of the MHO phenotype, which may evolve — gradually and chronically — into a phenotype with a higher cardiovascular risk, the metabolically unhealthy obesity (MUO). Therefore, the objective of this study was to assess autonomic dysfunction in individuals with MHO.

MATERIALS AND METHODS

Study Design and population

The present study is associated with a broader observational study entitled “Evaluation of stress indicators, body composition, and metabolic profile in employees of a cardiology reference hospital: contributions to quality of life—the Worker Health Study (ESAT)”, conducted at the National Institute of Cardiology (NIC), from November, 2018 to March, 2020. The inclusion criteria of ESAT study comprised: employees of NIC aged ≥ 18 years. The exclusion criteria were: medical sick leave, those assigned to another healthcare unit, undergone recent surgery, fasting for more than 13 h, being pregnant and/or lactating, not responding to the team’s attempts to contact for data collection (minimum of three attempts), or undergoing the second day of collection (D2) after more than two months from the first day of collection (D1). The ESAT study resulted in a comprehensive database, used in the present study, as a convenient sample size (Araujo et al. 2023, Carvalho et al. 2024). For the present study, individuals categorized as underweight (BMI≤18.5 kg/m2), unhealthy eutrophics (BMI>18.5 and <25.0 kg/m2) or overweight (BMI≥25.0 and <30.0 kg/m2) were excluded from the study. The protocol and all procedures were in accordance with the Helsinki Declaration revised in 2013 and the research ethics committee of National Institute of Cardiology approved the study with the following approved ethical protocol 96222718.7.0000.5272. All participants signed an informed consent form before enrolled the study.

Recruitment and Data Collection

After signing an informed consent, participants’ identification variables such as sex and age, along with clinical variables including self-reported chronic diseases, and regular medications in use were collected. The short version of the International Physical Activity Questionnaire (IPAQ-SF; Consisting of 7 questions assessing the frequency, time and intensity of physical activity performed at work, leisure time, commuting and household, and the time seated on a typical week and weekend day,) was used to evaluate physical activity levels, allowing individuals to be classified into three different categories: high, moderate and low physical activity levels (Matsudo et al. 2001, Hagströmer et al. 2006). This comprehensive data was acquired during a face-to-face interview conducted on the first day (D1) of the study. Subsequently, participants were requested to return for a second day (D2) dedicated to blood collection, anthropometric assessment, and autonomic function assessment. The time lapse between the initial assessment (D1) and the subsequent evaluations on the second day (D2) did not exceed 30 days.

Blood Sample Collection

Blood collection was performed after a 12-hour fasting. The serum glucose was evaluated using the hexokinase method. C-reactive protein was analyzed by turbidimetric immunoassay. Total cholesterol (TC), high-density lipoprotein cholesterol (HDL-c), low-density lipoprotein cholesterol (LDL-c) and triglycerides (TG) were evaluated by enzymatic colorimetric method. All biochemical analyses were performed using commercial kits and an automated method (ARCHITECT ci8200, Abbott ARCHIECT®, Abbott Park, IL, USA).

Anthropometric Measures

Body mass (kg) and height (m) were measured using a P-200C balance with stadiometer (Lider, Brazil). The Body Mass Index (BMI) was calculated by dividing the body mass by the square of the height and was classified according to the following BMI ranges: underweight (<18.5 kg/m²), eutrophic (18.5-24.9 kg/m²), overweight (25-29.9 kg/m²), and obesity (≥30 kg/m²) (Madden & Smith 2016).

Classification of anthropometric-metabolic profiles

After anthropometric classification in eutrophic or obesity, the volunteers were classified in metabolic profiles according to the International Diabetes Federation (IDF) criteria for metabolic syndrome. The items included in IDF criteria are (I) raised triglycerides (≥150 mg/dL or ≥1,7mmol/L) or specific treatment for this lipid abnormality; (II) reduced HDL cholesterol (<40 mg/dL or <1,04 mmol/L in males; and <50 mg/dL or <1,3 mmol/L in females) or specific treatment for this lipid abnormality; (III) raised blood pressure (blood pressure ≥ 130/85 mmHg) or treatment of previously identified hypertension; and (IV) raised fasting plasma glucose (≥100 mg/dL or ≥ 5,6 mmol/L) or previously diagnosed type 2 diabetes mellitus (Chiheb et al. 2016).

Participants that met one or more of the IDF criteria were considered metabolically unhealthy. Participants who did not have any of the IDF metabolic syndrome criteria were classified as metabolically healthy (Chiheb et al. 2016). These cutoff points, excluding abdominal circumference due its collinearity with BMI, were defined according to the proposed criteria for harmonized definitions of MHO reported by Lavie et al. (2018). Thus, the participants were classified in three different anthropometric-metabolic profiles: metabolically healthy eutrophic (MHE), MHO and MUO.

Autonomic Function Evaluation Protocol

Autonomic evaluation was performed after a 12-hour fast. The volunteers were instructed to stay in dorsal decubitus position for 15 minutes with head elevation from 30 degrees. Thereafter, they were instructed to get up and stay in orthostatic position for 3 minutes.

During the entire protocol execution time (18 minutes) the volunteers were monitored by the ECGV6® portable electrocardiograph (HW Sistemas, HeartWare Ltda), Polar® heart rate sensor (V800) and systemic blood pressure measurements (digital sphygmomanometer, G-Tech ®) at zero (T0), fifteen (T15) and eighteen minutes (T18) of the protocol.

The assessment of autonomic function was conducted in the morning (7:00-11:30 am) within a quiet room with a mild, controlled temperature (20-23°C/68-73°F). Throughout the evaluation, participants were instructed to maintain silence and remain at rest. Based on this protocol, the electrocardiogram, heart rate variability (HRV), autonomic test (30:15 ratio) and the orthostatic hypotension test were performed.

Autonomic Test (30:15 ratio)

The autonomic test, commonly known as the 30:15 Ratio or Ewing Ratio, was computed immediately following the commencement of the volunteer’s orthostatic period. The 30:15 ratio was determined by dividing the RR interval around the 15th heart beat (representing the minimum RR interval) by the RR interval around the 30th heart beat (representing the maximum RR interval), as assessed using the electrocardiogram DII lead. A normal result for this ratio is considered to be at least 1.04. Values below 1.04 were classified as abnormal (Rolim et al. 2008, Zygmunt & Stanczyk 2010).

Orthostatic Hypotension Test

The Orthostatic Hypotension Test, also known as the Postural Hypotension Test, was performed based on the variation (∆) between the blood pressure measurement at T15 and T18. Systolic blood pressure reduction (SBP) ≥20 mmHg and/or diastolic blood pressure reduction (DBP) ≥10 mmHg were considered abnormal, while SBP reduction between 10-19 mmHg and/or DBP between 5-9 mmHg were considered borderline (Lanier et al. 2011, Rolim et al. 2008, Weiss et al. 2004). All volunteers were asked about the occurrence of symptoms related to orthostatic hypotension such as dizziness, visual disturbances or pre-syncope (Weiss et al. 2004).

Heart Rate Variability (HRV)

RR intervals were obtained using Polar V800® heart rate sensor, and a 10-minute interference-free period tachogram (verified by electrocardiogram) analyzed with Kubios® software (v. 2.2, UEF, Finland). HRV indices in the frequency domain (calcutated analyzed by the Fast Fourier Transform method) were: absolute (ms2) and relative power (%), and peak frequency (Hz) of very low frequency (VLF; 0.0033-0.04 Hz), low frequency (LF; 0.04-0.15 Hz) and high frequency (HF; 0.15-0.4 Hz), also, the ratio of LF-to-HF power (LF/HF), and Total Power (Hz). In the time domain, the indices evaluated were mean of all normal RR intervals (MNN), Standard deviation of all normal RR intervals (SDNN), root-mean square of differences between adjacent normal RR intervals (rMSSD), number of adjacent RR intervals with a difference of duration greater than 50ms (NN50), percentage of adjacent RR intervals with a difference of duration greater than 50ms (pNN50), the integral of the density of the RR interval histogram divided by its height (triangular index) and Baseline width of the RR interval histogram (TINN). In addition, we also evaluated the nonlinear parameters of the Poincaré Plot: standard deviation perpendicular the line of identity (SD1), standard deviation along the line of identity (SD2) and the ratio of SD1-to-SD2 (SD1/SD2) ) (Task Force of the European Society of Cardiology & the North American Society of Pacing and Electrophysiology 1996, Shaffer & Ginsberg 2017, Gąsior et al. 2018).

Blinding and quality control of data collection

All data obtained during the study were collected through physical forms. Subsequently, the data were allocated in an online platform (Research Electronic Data Capture -REDCap) by an assistant researcher. A second assistant researcher checked all data entered in REDCap, correcting any errors. The anthropometric-metabolic profiles classification was performed only after analyzing the autonomic function of all volunteers, in order to maintain the blinding of the study and reduce possible bias.

Data Analysis and Statistics

Results are presented as mean ± standard deviation for continuous variables and as a number (percentage) for categorical variables. The Shapiro-Wilk test was used to assess the distribution of variables. Comparisons between the MHE, MHO and MUO groups were performed using One-way ANOVA with Sidak post-test for variables with normal distribution. Kruskal Wallis with Dunns post-test was used for variables with non-normal distribution. Categorical variables analysis was performed using Fisher’s exact test. Changes on blood pressure during the autonomic assessment protocol was performed using Two Way ANOVA for repeated measurements and Tukey’s post-test. Biochemical and autonomic assessment variables were also analyzed using sex and age-adjusted linear regression. The association of metabolic phenotype and biochemical and autonomic variables were evaluated using linear regression adjusted by sex and age. Linear regression was adjusted for sex and age because these factors can influence autonomic evaluation parameters (Voss et al. 2015). The significance level was set as P< 0.05. Statistical analyses and graphs were performed using STATA 16 software (Stata Corp USA) and GraphPad Prism 8.0.1 software (GraphPad Software, San Diego, California, USA), respectively.

RESULTS

Out of the initial 241 recruited volunteers for the study, eleven declined to participate. Following face-to-face interviews, 21 individuals met the exclusion criteria, and nine participants did not complete the assessments scheduled for the second day (D2), being consequently excluded from the study. From remaining 200, 102 were excluded because anthropometric-metabolic profile. Obesity prevalence was 38% (N=75) of total sample. Of these, 14 (7%) were MHO and 61 (31%) were classified MUO (Figure 1).

Figure 1
Recruitment of volunteers in the Worker’s Health Study (ESAT) and their categorization according to the anthropometric-metabolic profile.

Descriptive data of the present study total sample (n=98) and stratified by metabolic phenotype are shown in Table I. The mean age of the total sample was 43.8 years and BMI was approximately 32 kg/m², mostly women (65.3%). Hypertension was the main chronic disease self-reported only in MUO individuals. The antihypertensive drugs (angiotensin converting enzyme inhibitor - ACEi and angiotensin II type 1 receptor blocker – ARB) were the main medications in regular use were, self-reported only in MUO. The MUO group exhibited higher serum levels of glucose (p<0.001), triglycerides (p<0.001), LDL cholesterol (p=0.04) and C-reactive protein (p<0.001), and lower level of HDL cholesterol (P<0.001) compared to MHE. MHO presented higher level of HDL (P<0.001) and a lower level of triglycerides (p=0.007) compared to MUO. In addition, compared to MHE, MHO showed higher values of LDL cholesterol (P=0.01) and C-reactive protein (p=0.03). There was no difference in physical in physical activity level between groups.

Table I
Descriptive data of study’s participants.

VARIABLE	MHE (n=23) Mean ± SD or N (%)	MHO (n=14) Mean ± SD or N (%)	MUO (n=61) Mean ± SD or N (%)
Age (years)	40.2 ± 12.2	46.2 ± 8.8	44.7 ± 10.9
Gender N (%)
Women	15 (23.4%)	9 (14.1%)	40 (62.5%)
Men	8 (23.5%)	5 (14.7%)	21 (61.8%)
Weight (Kg)	63.2 ± 10.9	91.5 ± 12.4***	94.3 ± 16.5***
Height (cm)	1.66 ± 0.1	1.64 ± 0.1	1.64 ± 0.1
BMI (Kg/m2)	22.7 ± 1.3	33.5 ± 2.8***	34.7 ± 4.2***
Comorbidities
Hypertension	0 (0)	0 (0)	25 (41.0%)***
DM2	0 (0)	0 (0)	4 (4.1%)
Dyslipidemia	0 (0)	0 (0)	3 (4.9%)
hypothyroidism	0 (0)	1 (7.1%)	2 (3.1%)
Medications
β-blocker	0 (0)	0 (0)	3 (4.9%)
ACEi/ARB	0 (0)	0 (0)	16 (26.2%)***
Calcium channel blocker	0 (0)	0 (0)	2 (3.3%)
Antidiabetic	0 (0)	0 (0)	5 (8.2%)
Hypolipemiant	0 (0)	0 (0)	2 (3.3%)
Diuretics	0 (0)	0 (0)	9 (14%)
Physical activity level
Low	7 (30.4)	6 (46.1)	19 (31.1)
Moderate	8 (34.8)	1 (7.7)	15 (24.6)
High	8 (34.8)	6 (46.1)	27 (44.3)
Glucose (mg/dL)	84.5 ± 7.3	87.9 ± 4.7	99.9 ± 29.4***
Triglycerides (mg/dL)	74.9 ± 25.4	97.4 ± 34.3††	156.3 ± 93.6***
LDL-cholesterol (mg/dL)	116.8 ± 31.9	140.6 ± 40.8*	129.0 ± 31.8*
HDL-cholesterol (mg/dL)	67.0 ± 16.7	57.5 ± 9.3†††	46.3 ± 11.4***
Total cholesterol (mg/dL)	181.5 ± 30.7	198.1 ± 42.4	188.8 ± 38.4
C-Reactive Protein (mg/dL)	0.2 ± 0.4	0.3 ± 0.2*	0.6 ± 0.7***

• MHE= metabolically healthy eutrophic; MHO= metabolically healthy obesity; MUO= metabolically unhealthy obesity; BMI= body mass index; ACEi= angiotensin converting enzyme inhibitor; ARB= angiotensin II type 1 receptor blocker; LDL-cholesterol= low-density lipoprotein cholesterol. HDL-cholesterol= high-density lipoprotein cholesterol. Vs. MHE: *P<0.05; **P<0.01; ***P<0.001. Vs. MUO: †P<0.05; ††P<0.01; †††P<0.001.

Table II shows parameters related to autonomic function. MHO group presented lower heart rate (p=0.007) than MUO. QT interval duration was longer in MHO group compared to MHE (p=0.02) and MUO (p=0.007). However, after adjusting QT for heart rate using Bazett’s formula (QTc), only the MUO group presented longer QTc interval (p=0.03) compared to MHE.

Table II
Autonomic evaluation parameters in MHE, MHO and MUO groups.

VARIABLE	MHE (N=23) Mean ± SD or N (%)	MHO (N=14) Mean ± SD or N (%)	MUO (N=61) Mean ± SD or N (%)
ELECTROCARDIOGRAPHIC PARAMETERS
Heart Rate (bpm)	62.2 ± 9.3	58.6 ± 6.8††	66.4 ± 8.4
P wave voltage (mV)	1.6 ± 0.5	1.6 ± 0.5	1.6 ± 0.5
P wave duration (ms)	103.3 ± 8.8	108.5 ± 9.5	105.1 ± 12.6
T wave voltage (mV)	3.9 ± 1.4	3.7 ± 1.0	3.5 ± 1.4
QRS complex duration (ms)	82.9 ± 13.7	84.5 ± 23.9	84.3 ± 14.5
QT interval (ms)	402.7 ± 28.6	421.4 ± 23.9*††	403.1 ± 30.7
QTc interval (ms)	407.6 ± 21.7	415.3 ± 22.2	421.9 ± 22.8*
PR interval duration (ms)	160.2 ± 21.4	156.0 ± 19.8	165.6 ± 40.8
PR segment duration (ms)	56.6 ± 19.9	47.4 ± 19.8	55.6 ± 20.8
AUTONOMIC TEST
30th RR interval	889.04 ± 160.67	846.84 ± 149.25	799.31 ± 154.59
15th RR interval	766.52 ± 162.57	766.07 ± 142.66	696.18 ± 135.19
30:15 ratio	1.18 ± 0.2	1.12 ± 0.18	1.16 ± 0.18
30:15 ratio categorized (<1.04)	7 (30.43%)	7 (53.85%)	17 (29.31%)
ORTHOSTATIC HYPOTENSION TEST
SBP T0	116.56 ± 10.03	119.64 ± 11.57	131.0 ± 16.21
DBP T0	75.69 ± 7.51	75.64 ± 6.55	83.80 ± 10.93
SBP T15	114.0 ± 9.90	119.14 ± 6.35	127.88 ± 15.85
DBP T15	72.21 ± 6.76	76.64 ± 4.68	83.31 ± 10.44
SBP T18	116.21 ± 9.35	126.78 ± 10.59	133.08 ± 16.96
DBP T18	78.43 ± 7.88	85.14 ± 8.82	90.14 ± 10.92
SBP categorized (≥20 mmHg)	0 (0%)	0 (0%)	1 (1.64%)
DBP categorized (≥10 mmHg)	0 (0%)	0 (0%)	2 (3.28%)
HEART RATE VARIABILITY
Time Domain
MNN (ms)	985.3 ± 146.3	1005.7 ± 98.7†	915.5 ± 119.4
SDNN (ms)	58.9 ± 23.6	68.8 ± 29.2††	47.7 ±20.9*
rMSSD (ms)	59.7 ± 35.4	61.0 ± 39.6†	38.9 ± 21.9**
NN50 (count)	145.6 ± 107.1	154.6 ± 110.8†	102.0 ± 100.8*
pNN50 (%)	28.2 ± 23.4	29.8 ± 21.3†	18.4 ± 18.0*
RR triangular index	14371.9 ± 4339.5	14618.9 ± 7127.7	12242.6 ± 4899.6*
TINN (ms)	167.5 ± 115.6	172.5 ± 82.3	171.8 ± 9.1
Frequency Domain
VLF (Hz)	0.009 ± 0.008	0.01 ± 0.01	0.008 ± 0.006
VLF (%)	35.4 ± 18.8	43.0 ± 17.6	42.6 ± 17.5*
VLF (ms2)	952.5 ± 584.1	1628.9 ± 905.5*†††	936.6 ± 882.8
LF (Hz)	0.08 ± 0.02	0.07 ± 0.03	0.07 ± 0.03
LF (%)	27.1 ± 11.4	29.2 ± 11.2	28.0 ± 12.4
LF (nu)	44.7 ± 18.9	52.9 ± 18.2	50.5 ± 18.4
LF (ms2)	947.7 ± 905.7	1632.6 ± 1847.5††	709.6 ± 850.9
HF (Hz)	0.24 ± 0.05	0.25 ± 0.06	0.2 ± 0.06
HF (%)	37.4 ± 20.0	27.7 ± 14.9	29.3 ± 15.3
HF (nu)	55.12 ± 18.89	47.0 ± 18.1	49.3 ± 18.4
HF (ms2)	1534.4 ± 1606.2	1899.7 ± 2648.7*†	718.7 ± 830.4
Total Power (ms2)	3438.3 ± 2662.3	5163.2 ± 4743.8††	2366.6 ± 2189.7*
Sympathovagal Index (LF/HF)	1.05 ± 0.77	1.66 ± 1.75	1.52 ± 1.61
LF/HF categorized	19 (82.6%)	13 (92.8%)	52 (85.2%)
Non-linear Indexes
SD1 (ms)	42.3 ± 25.1	43.2 ± 28.0†	27.6 ± 15.5**
SD2 (ms)	70.7 ± 25.6	86.7 ± 32.0††	60.9 ± 26.5
SD1/SD2	0.6 ± 0.2	0.5 ± 0.1	0.4 ± 0.1*

• MHE= metabolically healthy eutrophic; MHO= metabolically healthy obesity; MUO= metabolically unhealthy obesity; SBP= systolic blood pressure; DBP= diastolic blood pressure; QTc= QT corrected by bazett’s formula (QTc=QT/√RR). Analysis with One Way ANOVA and Sidak post-test or Kruskal Wallis and Dunn’s post-test. Vs. MHE: *P<0.05; **P<0.01; ***P<0.001. Vs. MUO: †P<0.05; ††P<0.01; †††P<0.001.

The borderline change in SBP (10-19 mmHg reduction) was not presented by any of the three groups, while the borderline change in DBP (5-9 mmHg reduction) was presented by 1 (4.35%) of the MHE, none of MHO and 1 (1.64%) of the MUO group, without significant difference (p= 0.856).

There were no differences in time domain indexes of HRV comparing MHO and MHE, but Compared to MUO, MHO group exhibited higher MNN (p=0.04), SDNN (p=0.003), rMSSD (p=0.02), NN50 (p=0.03) and pNN50 (p=0.02). MUO showed lower values of SDNN (p=0.02), rMSSD (p=0.008), NN50 (p=0.03), pNN50 (p=0.03) and triangular index (p =0.01) compared to MHE.

Observing the frequency domain indexes of HRV, MHO presented higher VLF (ms²) than MHE (p=0.016) and MUO (p<0.001) groups. Similarly, MHO shown higher HF (ms²) compared to MHE (p=0.012) and MUO (p=0.039). In addition, MHO group presented higher LF (ms²) compared to the MUO (p=0.005). MUO had a higher VLF (%) (p=0.04) than MHE. In addition, MHO group presented higher (p=0.002) Total Power compared to MUO, while MUO exhibited lower (p=0.02) Total Power compared to MHE group. Regarding the non-linear indexes, MHO presented higher SD2 index (p=0.002) to MUO, while MUO presented lower SD1 index compared to the MHE (p=0.008) and the MHO (p=0.020). Finally, comparing MHO and MHE, there was no difference in SD1/SD2 ratio, but the MUO group presented lower (p=0.010) SD1/SD2 ratio compared to the MHE. The other electrocardiogram and HRV parameters showed no statistically significant difference between groups.

Figure 2 presents SBP and DBP variation during the orthostatic hypotension test. There was no statistical difference in systolic and diastolic blood pressures comparing MHO (119.6 ± 11.6 and 75.6 ± 6.5, SBP and DBP respectively) and MHE (116.6 ± 10.0 and 75.7 ± 7.5 mmHg, SBP and DBP respectively) groups at baseline, but also comparing MHO (119.1 ± 6.3 and 76.6 ± 4.7 mmHg, SBP and DBP respectively) and MHE (114.0 ± 9.9 and 72.2 ± 6.8 mmHg, SBP and DBP respectively) at the end of the 15 minutes resting period. At baseline, the mean MUO SBP (131.0 ± 16.2 mmHg) and DBP (83.8 ± 10.9 mmHg) were higher than observed in MHE (P<0.001) and MHO (P<0.05 and P<0.01, respectively). In the end of resting period (15 minutes) the MUO blood pressure (127.9 ± 15.8 and 83.3 ± 10.4 mmHg, SBP and DBP respectively) were higher than observed in MHE (P<0.001) and MHO (P<0.01). After orthostasis, the SBP of MHO group presented higher (P=0.0135) than observed in MHE group (126.78 ± 10.6 vs. 116.21 ± 9.3 mmHg, respectively), but not different from the MUO group (133.08 ± 16.96 mmHg), that similarly exhibited higher SBP compared to MHE (P<0.0001). Despite DBP of MUO (90.1 ± 10.9 mmHg) was higher (P<0.0001) than observed in MHE (78.4 ± 7.9 mmHg), there was no difference between MHO comparing both MHE and MUO.

Figure 2
Blood pressure fluctuations during the autonomic evaluation protocol. Variation in systolic blood pressure (SBP, mmHg), in a, and diastolic blood pressure (DBP, mmHg), in b, throughout the autonomic evaluation protocol at 0 (beginning of rest), 15 (end of rest) and 18 (end of orthostasis) minutes. Two Way ANOVA and Tuckey post-test. MHE= metabolically healthy eutrophic; MHO= metabolically healthy obesity; MUO= metabolically unhealthy obesity; BMI= body mass index. Vs. MHE: *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001. Vs. MUO: †P<0.05; ††P<0.01; †††P<0.001.

Table III shows the association between metabolic phenotypes and outcomes adjusted by sex and age. MUO was associated with lower HDL cholesterol (p=0.001), MNN (p=0.003), rMSSD (p=0.011), SD1 (p=0.011) and SD1/SD2 ratio (p=0.012), and higher glucose (p=0.018), triglycerides (p=0.001), C-reactive protein (p=0.005), QTc interval (p=0.004) compared to MHE group. However, the MHO was independently associated with lower HDL cholesterol (p=0.009), and higher VLF (ms²) (p=0.018), LF (ms²) (p=0.028), Total Power (p=0.023) and SD2 (p=0.037) HRV parameters compared to MHE group.

Table III
Gender- and age-adjusted simple linear regression using the anthropometric-metabolic profile as dependent variable.

VARIABLE	MHO (N=14)			MUO (N=61)
	β	CI 95%	P - VALUE	β	CI 95%	P - VALUE
METABOLIC PARAMETERS
Glucose (mg/dL)	+1.8	-14.4 to +18.1	0.823	+14.3	+2.6 to +26.0	0.018
Total-cholesterol (mg/dL)	+13.8	-11.7 to +39.4	0.286	+5.3	-13.2 to +23.7	0.572
HDL-cholesterol (mg/dL)	-10.7	-18.6 to -2.8	0.009	-21.7	-27.4 to -16.0	0.001
LDL-cholesterol (mg/dl)	+22.6	-0.3 to +45.5	0.053	+11.3	-5.2 to +27.8	0.177
Triglycerides (mg/dL)	+15.6	-35.9 to +67.2	0.549	+76.6	+39.4 to +113.9	0.001
C-Reactive Protein (mg/dL)	+0.1	-0.3 to +0.5	0.564	+0.4	+0.1 to +0.7	0.005
ORTHOSTATIC TEST
30:15 Ratio	-0.03	-0.2 to +0.1	0.602	+0.001	-0.1 to +0.1	0.978
ORTHOSTATIC HYPOTENSION TEST
Delta SBP	+4.4	-0.9 to +9.8	0.105	+2.2	-1.7 to +6.1	0.262
Delta DBP	+2.6	-1.9 to +7.0	0.255	+0.9	-2.3 to +4.1	0.594
ECG
QT interval (ms)	+14.1	-4.9 to +33.1	0.144	-3.2	-16.9 to +10.5	0.647
QTc interval (ms)	+9.7	-4.7 to +24.1	0.186	+15.5	+5.1 to +26.0	0.004
HRV
MNN (ms)	-4.8	-83.9 to +74.3	0.904	-87.6	-145.1 to -30.1	0.003
SDNN (ms)	+13.0	-1.8 to +27.9	0.085	-9.0	-19.5 to +1.8	0.100
rMSSD (ms)	+6.3	-11.9 to +24.5	0.495	-17.2	-30.5 to -4.0	0.011
NN50 (count)	+32.7	-32.1 to +97.6	0.319	-26.8	-73.9 to +20.4	0.262
pNN50 (%)	+5.4	-7.4 to +18.2	0.403	-7.1	-16.40 to +2.2	0.133
TINN (ms)	+32.4	-24.2 to +89.1	0.258	+23.8	-17.3 to +65.0	0.254
RR triangular index	+772.9	-2705.9 to +4251.7	0.660	-1755.4	-4284.2 to +773.3	0.171
VLF (Hz)	+0.003	-0.001 to +0.009	0.186	-0.0002	-0.004 to +0.003	0.908
VLF (ms2)	+677.1	+118.9 to +1235.4	0.018	-14.7	-420.5 to +391.1	0.943
VLF (%)	+4.2	-7.2 to +15.7	0.462	+4.8	-3.5 to +13.1	0.258
LF (Hz)	+0.001	-0.01 to +0.02	0.886	-0.0004	-0.01 to +0.01	0.955
LF (ms2)	+772.0	+87.4 to +1456.6	0.028	-175.5	-673.1 to +322.1	0.485
LF (%)	+1.8	-6.1 to +9.6	0.655	+0.7	-5.0 to +6.4	0.808
LF (nu)	+5.1	-6.6 to +16.9	0.390	+3.6	-4.9 to +12.2	0.399
HF (Hz)	+0.009	-0.03 to +0.05	0.648	+0.01	-0.01 to +0.04	0.479
HF (ms2)	+618.9	-269.8 to +1507.7	0.170	-635.3	-1281.3 to +10.7	0.054
HF (%)	-6.0	-16.23 to +4.3	0.250	-5.4	-12.9 to +2.0	0.151
HF (nu)	-5.1	-16.8 to +6.7	0.394	-3.6	-12.2 to +4.9	0.398
Total Power (ms2)	+2067.2	+292.2 to +3842.3	0.023	-827.0	-2117.3 to +463.3	0.206
LF/HF ratio	+0.4	-0.5 to +1.4	0.388	+0.3	-0.4 to +1.1	0.335
SD1 (ms)	+4.4	-8.4 to +17.3	0.495	-12.2	-21.6 to -2.8	0.011
SD2 (ms)	+19.1	+1.2 to +36.9	0.037	-7.5	-20.5 to +5.5	0.254
SD1/SD2 ratio	-0.07	-0.2 to 0.0	0.144	-0.09	-0.2 to 0.0	0.012

• Gender- and age-adjusted simple linear regression analysis. CI= confidence interval; MHE= metabolically healthy eutrophic; MHO= metabolically healthy obesity; MUO= metabolically unhealthy obesity; LDL-cholesterol= low-density lipoprotein cholesterol; LF/HF= Sympathovagal Index; HDL- cholesterol= high-density lipoprotein cholesterol; SBP= systolic blood pressure; DBP= diastolic blood pressure; QTc= QT corrected by bazett’s formula (QTC=QT/√RR).

DISCUSSION

Our study represents a pioneering effort in assessing autonomic function among individuals with MHO, employing a comprehensive array of methods including HRV analysis. The key findings of our investigation were: (1) a low prevalence of MHO within our cohort, (2) discernible alterations in time and frequency domain parameters of HRV in MHO, and (3) notable alteration in systolic blood pressure response during the orthostatic hypotension test among individuals classified as MHO.

The present study evaluated the prevalence of obesity and MHO in a total sample of 200 volunteers. Most volunteers recruited in the present study were women (65.3%) and the mean age was 43.8 years. Rastović et al. (2019) obtained a similar profile in a study with 99 individuals with obesity that also evaluated HRV in different obesity phenotypes, whose average age was 41.28 years and the prevalence of women was 63.6%.

According to the World Health Organization, the prevalence of obesity in adults (aged 18 years or older) is 13% (WHO 2022). In 2025, the estimated worldwide prevalence of obesity will increase to 18% in men and exceed 21% in women (NCD-RisC 2016). In the present study, we obtained 38% of obesity prevalence, a higher rate than 30-34% obesity prevalence in United States between 2005 and 2015 (Chooi et al. 2019).

MHO, a so-called “benign” obesity phenotype, has different prevalence according to the criteria used for its definition (Velho et al. 2010, Liu et al. 2019). In the present study, we used the absence of IDF metabolic syndrome criteria to define MHO classification. Abdominal circumference was excluded due to its collinearity with BMI. These cutoff points were based on the harmonized definition for MHO criteria (Lavie et al. 2018). We obtained 7% of MHO prevalence in the total sample (n=200) and 18.7% of MHO prevalence in subjects with obesity (N=75). Wildman et al. (2008), in the National Health and Nutrition Examination Surveys (NHANES) study, carried out between 1999-2004, demonstrated slightly higher prevalence. They used cutoff points of IDF metabolic syndrome criteria associated to inflammation and insulin resistance markers to define MHO profile. Wildman et al. (2008) found 9.7% of MHO prevalence among the US adult population and 31.7% among subjects with obesity of this population.

Similar to that found in our study, a population-based cohort study named Risk Evaluation of Cancers in Chinese Diabetic Individuals: a longitudinal study (REACTION), carried out between 2011 and 2015, showed 6.7% of MHO prevalence among the total sample and 26.1% of MHO prevalence among individuals with obesity. However, despite the MHO classification cutoff points having been similar to those used in our study, Li et al. (2019) classified as metabolically healthy who had less than two criteria.

Despite the “metabolically healthy” status, individuals with MHO showed changes in biochemical parameters, such as higher levels of LDL cholesterol and C-reactive protein, compared to MHE. Manu et al. (2012) previously demonstrated similar results, with increased levels of non-HDL cholesterol and C-reactive protein in MHO, but without significant LDL cholesterol alterations compared to MHE. According to the REACTION cohort, elevated LDL levels are an independent risk predictor of transition from MHO to the metabolically unhealthy phenotype, whereas low LDL levels suggest a propensity to maintain the same metabolic phenotype and are associated with lower cardiovascular risk (Li et al. 2019).

When evaluating the HRV, we noticed that the MHO group presented a higher VLF index (ms²) in comparison to MHE and MUO in the variance and linear regression analysis. Fujibayashi et al. (2009) demonstrated a positive independent association between the VLF variation (ms²) and the HDL/CT ratio in individuals with obesity (Fujibayashi et al. 2009), which may explain the finding of an increase in this index in the MHO group, since they did not present a reduction in HDL-cholesterol.

In contrast, low levels of the VLF component are associated with high levels of chronic inflammation (Usui & Nishida 2017). In addition, VLF reduction is more strongly associated with all-cause mortality than the LF and HF indices reduction (Shaffer & Ginsberg 2017). VLF fluctuations can be explained due to several physiological mechanisms such as thermoregulation, breathing pattern, parasympathetic activity and renin-angiotensin-aldosterone system activation (Guzzetti et al. 2005). Therefore, one of our study hypotheses is that the increase in VLF index represents greater parasympathetic activation in MHO group.

In the present study, MHO also showed a higher level of the HF index (ms²) compared to MHE and MUO. However, in the multivariate analysis this result does not remain statistically significant. Yadav et al. (2017) demonstrated that individuals with obesity had a lower HF mean (ms²) compared to eutrophic individuals. Similarly, overweight individuals also demonstrated a reduction in HF (nu) compared to eutrophic individuals (Chintala et al. 2015). The HF index is associated with the activity of the parasympathetic nervous system. Thus, during increased vagal tone, there is a propensity to increase HF even as other components of HRV (Reyes del Paso et al. 2013).

On the other hand, conditions associated with chronic sympathetic activation promote a propensity to reduce HRV parameters, including the LF index (Reyes del Paso et al. 2013). In our study, the MHO group showed a higher LF index (ms²) in comparison to the MUO in the variance analysis while, in the regression analysis, there is an independently association between this index and MHO profile. The LF index is associated with sympathetic and parasympathetic modulation, but predominantly reflects sympathetic activity (Ernst 2017). In the study conducted by Yadav et al. (2017), individuals with obesity showed a reduction in LF (ms²) in comparison to eutrophic individuals and did not present a significant difference in LF (nu). In overweight subjects, according to the Chintala et al. (2015) there is a significant increase in the LF (nu) index compared to eutrophic subjects.

Even in the frequency domain, the Total Power (ms²) represents the sum of all the previously mentioned components (VLF, LF, and HF), receiving influence from the sympathetic and parasympathetic activity (Task Force of the European Society of Cardiology & the North American Society of Pacing and Electrophysiology 1996, Voss et al. 2015, Shaffer & Ginsberg 2017, Gąsior et al. 2018, Rastović et al. 2019). In the present study, the MHO group has a higher level of Total Power compared to the MUO in the variance and linear regression analysis. In previous study, a reduction in Total Power was demonstrated in overweight subjects (Chintala et al. 2015). Furthermore, after linear regression adjusted for insulin dosage and the HOMA-IR index, MHO presented a significant increase in Total Power compared to MUO group, similar to the finding in our study (Rastović et al. 2016).

In HRV time domain, MUO group showed lower levels of SDNN, rMSSD, NN50, and pNN50 compared to the MHO and MHE. However, in the linear regression analysis, only the MNN and rMSSD indices are negatively associated with the MUO profile. MNN and SDNN indices represent sympathetic and parasympathetic activity, whereas the rMSSD, NN50, and pNN50 indices correspond to parasympathetic activity. In a previous study, MUO subjects also had lower rMSSD levels, as well as SDNN and pNN50 indices after linear regression adjusted for BMI, insulin dosage, and HOMA-IR (Rastović et al. 2016).

In the HRV non-linear indices, MHO has a higher level of the SD2 parameter compared to the MUO group. Equally, SD2 index is independently and positively associated to MHO profile. Nevertheless, SD1 index and SD1/SD2 ratio are negatively associated to MUO profile. The SD1 index represents the parasympathetic activity, while the SD2 index represents the HRV in a global way, reflecting the sympathetic and parasympathetic activity. The SD1/SD2 ratio represents the variations that occur from the fast and slow adjustment to the heart rate, even as LF/HF representing the autonomic balance (Abreu et al. 2014, Shaffer & Ginsberg 2017). A significant reduction in the SD1 and SD2 indices has previously been demonstrated in children with obesity, but without a significant difference in the SD1/SD2 ratio (Vanderlei et al. 2010). In the study by Yadav et al. (2017), normotensive adults with obesity showed a significant reduction in the SD1 index compared to eutrophic individuals, without significant difference in the SD2 and SD1/SD2 indices.

Considering previous research, there are only two published studies evaluating HRV in different obesity profiles, both performed by Rastović et al. (2016). The first, published in 2016, evaluated HRV among 44 premenopausal women (19-51 years old) divided between the MHO and MUO groups. There was no difference in any HRV indices evaluated (MNN, SDNN, rMSSD, pNN50, LF ms², LF nu, HF ms², HF nu, Total Power, and LF/HF) using three different categorizations of “metabolically healthy” profile: IDF criteria, Wildman criteria, and isolated HOMA-IR insulin sensitivity criteria. However, when performing SBP-adjusted multivariate regression analysis, MHO group (according to the Wildman classification) showed a reduction in MNN index and an increase in LF/HF compared to MUO group (Rastović et al. 2016).

The second study, published in 2019, 99 subjects with obesity (both genders) aged 19-61 years were evaluated. Rastović et al. (2019) demonstrated that younger MHO individuals (19-29 years old) have a higher HF index (ms²), whereas in older MHO individuals (30-39, 40-49 and 50-59 years old) this index decreases significantly. In the MUO group, a significant reduction (approximately 30%) was also observed in the SDNN, rMSSD, lnpNN50, lnLF, lnHF, and Total Power indices with increasing age. However, this decrease in MUO group was maintained until the age of 40-49 years, stabilizing afterwards. These findings suggest that MUO group is more influenced by age on the HRV indices than MHO group (Rastović et al. 2019).

MHO subjects had a longer QT interval duration when compared to MUO and MHE group. QT interval represents the period from the beginning of ventricular depolarization until the end of its repolarization. Thus, QT interval prolongation may induce re-entry and provoke ventricular arrhythmias such as ventricular fibrillation and Torsades des Pointes, becoming an independent risk marker for sudden death (Straus et al. 2006). However, this alteration does not remain as a statistically significant result in the gender and age-adjusted linear regression analysis. In our study, MUO group had a longer QTc interval duration compared to MHE in the variance and linear regression analysis. Perhaps, this result could be explained by the presence of arterial hypertension, diabetes mellitus, dyslipidemia and abdominal obesity, which may also increase the risk of QTc prolongation (Van Noord et al. 2010, Ma et al. 2019).

The electrocardiogram also allowed evaluation of the 30:15 ratio, the most accurate of all Ewing tests. The 30:15 ratio is used to assess parasympathetic (cardiovagal) regulation and it is an independent predictor of CAN (Pafili et al. 2015). Furthermore, there is a strong association between increased BMI and increased risk of developing CAN, representing cardiovascular risk (Williams et al. 2019). This is the first study to evaluate 30:15 ratio between different obesity phenotypes, since there is still no study in the literature that evaluates the 30:15 function in MHO, only in individuals with obesity (Yakinci et al. 2000, Ali et al. 2016, Jain et al. 2019). Although MHO group showed a 30:15 ratio alteration prevalence greater than 50%, there was no significant difference in comparison to the other groups. Similar to our study finding, Ali et al. (2016) evaluated 30:15 ratio between adults with hypertension and obesity vs those with hypertension but without obesity, but did not obtain a statistically significant difference.

Another autonomic test performed in the present study was the orthostatic hypotension test, which showed no significant alteration between the MHE, MHO and MUO groups. Orthostatic hypotension is a disorder that increases in incidence with age and with comorbidities presence (as valve heart disease, diabetes, neurodegenerative diseases), which is associated to falling risk and increased morbidity and mortality (Shaw et al. 2017).

Nevertheless, we observed an unexpected behavior in pressure level variation curve in MHO group during the protocol execution time. MUO group has higher levels of SBP and DBP since protocol baseline and MHO group suddenly after orthostasis showed a significant increase in SBP, resembling MUO group. This behavior may explain the low hypotension orthostatic prevalence and characterize the phenomenon called “orthostatic hypertension”, which represents the sustained elevation of pressure after orthostasis (Robertson 2011).

One of the possible mechanisms that could explain orthostatic hypertension after orthostasis could be the venous volume accumulation in extremities followed by a cardiac output reduction and plasma norepinephrine increase (Robertson 2011). Orthostatic hypertension may be just an incidental finding or an autonomic dysfunction condition, with baroreflex dysfunction and excessive increase in sympathetic activity (Robertson 2011, Chhabra & Spodick 2013). This is an undervalued clinical entity and does not have strict diagnostic criteria, although a cutoff of 20 mmHg increase in systolic pressure is suggested as clinical diagnosis (Chhabra & Spodick 2013, Magkas et al. 2019).

Although not commonly associated with clinical symptoms, orthostatic hypertension can cause dizziness, headache, palpitations, nausea, sweating, and, more rarely, syncope (Kario 2013, Magkas et al. 2019). Even though this clinical condition does not commonly generate significant symptoms, orthostatic hypertension may increase cardiovascular risk and should be further investigated (Kario 2013). Nibouche-Hattab et al. (2017) suggested that orthostatic hypertension is related to metabolic syndrome and predicts hypertension in normotensive individuals recently diagnosed with type 2 diabetes mellitus. Hu et al. (2020), when assessing this clinical entity in children, suggested that increased BMI might be one of the risk factors for developing orthostatic hypertension. However, there are still no studies available in the literature evaluating orthostatic hypertension in obesity and its different phenotypes. Therefore, our findings and this clinical entity discovered in the present study should serve as a warning for further studies to investigate its potential cardiovascular risk, making it more recognized in clinical practice.

Limitations

In the present study, the Tilt Test and another Ewing’s Battery Test were not performed. These methodologies are part of autonomic dysfunction scope diagnosis, but were not available in our institution for research purposes.

However, this is the first study to perform two cardiovascular autonomic reflex tests (30:15 ratio and hypotension orthostatic test), associated to ECG and HRV for autonomic assessment in different obesity phenotypes (MHO and MUO). Considering previous research, there are only two articles (Rastović et al. 2016, 2019) evaluating autonomic function in these profiles, whose methodology was only HRV.

Despite the small sample size (n=98) subdivided into MHE (n=23), MHO (n=14) and MUO (n=61), with MHO prevalence of 7% (14 individuals out of 200 volunteers evaluated), the post-hoc analysis of the statistical power of the sample — carried out with G*Power version 3.1.9.2 — showed reasonable statistical power (0.86 for HF; 0.95 for VLF; 0.91 for LF and 0.99 for PASt18), corroborating the evidence that the total sample was sufficient to verify the findings obtained in the present study.

CONCLUSIONS

Our findings indicate that MHO subjects present alterations in HRV, mainly represented by the higher level of the linear indices VLF, HF, LF, and Total Power in the frequency domain and by the higher level of SD2 among the non-linear indices, suggesting increased parasympathetic activation. This increase in parasympathetic activation may represent an organic attempt to maintain homeostasis and do not manifest the metabolic syndrome alterations. These findings, associated with an unexpected increased SBP after orthostasis and higher serum levels of C-reactive protein and LDL cholesterol suggest that the presence of MHO, even in the absence of metabolic syndrome, predisposes to the occurrence of subclinical autonomic alterations, exposing these subjects to a higher risk of autonomic dysfunction and, consequently, elevated cardiovascular risk, suggesting that MHO is — in fact — a transition stage to the development of metabolically unhealthy obesity.

Acknowledgements

The Worker’s Health Study (ESAT) was supported by research grants from Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ) (E43/2021, E26/211.136/2021, and E-26/201.476/2021) and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) (424869/2021-7). The authors would like to thank the Worker’s Health Study (ESAT) investigators: F. G. Jesus, I. M. Barbosa, N. E. P. Andrade Junior, J. V. C. Mello. In addition, we would like to thank the volunteers’ participation and all support of National Institute of Cardiology (RJ, Brazil). The authors declare no competing interests.

• Data availability
  All the raw data will be made available in the link https://osf.io/kuxag/ after acceptation of the manuscript.

References

• ABREU LCD, ROQUE AL, CASTRO BCRD, SOUZA ACAD, VANDERLEI LCM, FERREIRA LL, RAIMUNDO RD, FONSECA FLA, VALENTI VE & CISTERNAS JR. 2014. Responses of the geometric indices of heart rate variability to the active orthostatic test in women. MedicalExpress 1: 351-355.
• AKHTER S, BEGUM N, FERDOUSI S & KHAN MSM. 2011. Autonomic Neuropathy in Obesity. J Bangladesh Soc Physiol 6: 9. https://doi.org/https://doi.org/10.3329/jbsp.v6i1.8059.
  » https://doi.org/10.3329/jbsp.v6i1.8059
• AKIL L & AHMAD HA. 2011. Relationships between obesity and cardiovascular diseases in four southern states and Colorado. J Health Care Poor Underserved 22: 61-72. https://doi.org/10.1353/hpu.2011.0166.
  » https://doi.org/10.1353/hpu.2011.0166
• ALI A, GANAI J, MUTHUKRISHNAN S & KOHLI S. 2016. Evaluation of Autonomic Dysfunction in Obese and Non-Obese Hypertensive Subjects. J Clin Diagn Res 10: YC01-YC03. https://doi.org/10.7860/jcdr/2016/18780.7923.
  » https://doi.org/10.7860/jcdr/2016/18780.7923
• ANDRES R. 1980. Effect of obesity on total mortality. Int J Obes 4: 381-386.
• ARAUJO CFDS, MELLO JVC, DUQUE AP, NOGUEIRA ICS, MEDIANO MFF, RODRIGUES JUNIOR LF & HUGUENIN GVB. 2023. Lack of association between metabolic phenotype and food consumption by degree of food processing: results from the Study of Workers’ Health (ESAT). Nutr Hosp 40: 119-127. https://doi.org/10.20960/nh.04242.
  » https://doi.org/10.20960/nh.04242
• BARROSO TA, MARINS LB, ALVES R, GONÇALVES ACS, BARROSO SG & ROCHA GDS. 2017. Association of Central Obesity with The Incidence of Cardiovascular Diseases and Risk Factors. Int J Cardiovasc Sci 30: 416-424.
• BOUTARI C & MANTZOROS CS. 2022. A 2022 update on the epidemiology of obesity and a call to action: as its twin COVID-19 pandemic appears to be receding, the obesity and dysmetabolism pandemic continues to rage on. Metabolism 133: 155217. https://doi.org/10.1016/j.metabol.2022.155217.
  » https://doi.org/10.1016/j.metabol.2022.155217
• CANALE MP, MANCA DI VILLAHERMOSA S, MARTINO G, ROVELLA V, NOCE A, DE LORENZO A & DI DANIELE N. 2013. Obesity-related metabolic syndrome: mechanisms of sympathetic overactivity. Int J Endocrinol 2013: 865965. https://doi.org/10.1155/2013/865965.
  » https://doi.org/10.1155/2013/865965
• CARVALHO MO, DUQUE AP, HUGUENIN GVB, FELIX MEDIANO MF & RODRIGUES JÚNIOR LF. 2024. Increased Cardiometabolic Risk in Dynapenic Obesity: Results from the Study of Workers’ Health (ESAT). Life 14: 1174. https://doi.org/10.3390/life14091174.
  » https://doi.org/10.3390/life14091174
• CHHABRA L & SPODICK DH. 2013. Orthostatic hypertension: recognizing an underappreciated clinical condition. Indian Heart J 65: 454-456. https://doi.org/10.1016/j.ihj.2013.06.023.
  » https://doi.org/10.1016/j.ihj.2013.06.023
• CHIHEB S, COSSON E, BANU I, HAMO-TCHATCHOUANG E, CUSSAC-PILLEGAND C, NGUYEN MT & VALENSI P. 2016. Are Obese Individuals with no Feature of Metabolic Syndrome but Increased Waist Circumference Really Healthy? A Cross Sectional Study. Exp Clin Endocrinol Diabetes 124: 410-416. https://doi.org/10.1055/s-0035-1569264.
  » https://doi.org/10.1055/s-0035-1569264
• CHINTALA KK, KRISHNA BH & N MR. 2015. Heart rate variability in overweight health care students: correlation with visceral fat. J Clin Diagn Res 9: CC06-CC08. https://doi.org/10.7860/JCDR/2015/12145.5434.
  » https://doi.org/10.7860/JCDR/2015/12145.5434
• CHOOI YC, DING C & MAGKOS F. 2019. The epidemiology of obesity. Metabolism 92: 6-10. https://doi.org/10.1016/j.metabol.2018.09.005.
  » https://doi.org/10.1016/j.metabol.2018.09.005
• CSIGE I, UJVÁROSY D, SZABÓ Z, LŐRINCZ I, PARAGH G, HARANGI M & SOMODI S. 2018. The Impact of Obesity on the Cardiovascular System. J Diabetes Res 2018: 3407306. https://doi.org/10.1155/2018/3407306.
  » https://doi.org/10.1155/2018/3407306
• EBONG IA, GOFF DC, RODRIGUEZ CJ, CHEN H & BERTONI AG. 2014. Mechanisms of heart failure in obesity. Obes Res Clin Pract 8: e540-e548. https://doi.org/10.1016/j.orcp.2013.12.005.
  » https://doi.org/10.1016/j.orcp.2013.12.005
• ERNST G. 2017. Heart-Rate Variability-More than Heart Beats? Front Public Health 5: 240. https://doi.org/10.3389/fpubh.2017.00240.
  » https://doi.org/10.3389/fpubh.2017.00240
• FUJIBAYASHI M ET AL. 2009. Thermoregulatory sympathetic nervous system activity and diet-induced waist-circumference reduction in obese Japanese women. Am J Hum Biol 21: 828-835. https://doi.org/10.1002/ajhb.20899.
  » https://doi.org/10.1002/ajhb.20899
• GĄSIOR JS, SACHA J, PAWŁOWSKI M, ZIELIŃSKI J, JELEŃ PJ, TOMIK A, KSIĄŻCZYK TM, WERNER B & DĄBROWSKI MJ. 2018. Normative Values for Heart Rate Variability Parameters in School-Aged Children: Simple Approach Considering Differences in Average Heart Rate. Front Physiol 9: 1495. https://doi.org/10.3389/fphys.2018.01495.
  » https://doi.org/10.3389/fphys.2018.01495
• GUARINO D, NANNIPIERI M, IERVASI G, TADDEI S & BRUNO RM. 2017. The Role of the Autonomic Nervous System in the Pathophysiology of Obesity. Front Physiol 8: 665. https://doi.org/10.3389/fphys.2017.00665.
  » https://doi.org/10.3389/fphys.2017.00665
• GUZZETTI S, LA ROVERE MT, PINNA GD, MAESTRI R, BORRONI E, PORTA A, MORTARA A & MALLIANI A. 2005. Different spectral components of 24 h heart rate variability are related to different modes of death in chronic heart failure. Eur Heart J 26: 357-362. https://doi.org/10.1093/eurheartj/ehi067.
  » https://doi.org/10.1093/eurheartj/ehi067
• HAGSTRÖMER M, OJA P & SJÖSTRÖM M. 2006. The International Physical Activity Questionnaire (IPAQ): a study of concurrent and construct validity. Public Health Nutr 9: 755-762. https://doi.org/10.1079/PHN2005898.
  » https://doi.org/10.1079/PHN2005898
• HATHTHOTUWA RN, WIJEYARATNE CN & SENARATH U. 2020. Worldwide epidemic of obesity. In: Mahmood TA et al. (Eds), Obesity and Obstetrics (Second Edition), Elsevier, p. 3-8.
• HU Y, HE B, HAN Z, WANG Y, TAO C, JIN H & DU J. 2020. Risk Factors for Orthostatic Hypertension in Children. J Pediatr 227: 212-217. https://doi.org/10.1016/j.jpeds.2020.07.030.
  » https://doi.org/10.1016/j.jpeds.2020.07.030
• JAIN NB, YADAV A, SHANKAR N & AHMED R. 2019. Correlation of C1q/TNF-Related Protein-3 with Cardiac Autonomic Tone and Metabolic Parameters in Obesity. Int J Endocrinol Metab 17: e74250. https://doi.org/10.5812/ijem.74250.
  » https://doi.org/10.5812/ijem.74250
• KARIO K. 2013. Orthostatic hypertension-a new haemodynamic cardiovascular risk factor. Nat Rev Nephrol 9: 726-738. https://doi.org/10.1038/nrneph.2013.224.
  » https://doi.org/10.1038/nrneph.2013.224
• KEYES A. 1973. Overweight and the risk of sudden heart attack and sudden death. In: Obesity in Perspective, DHEW Publication No (NIH) 75-708.
• LANIER JB, MOTE MB & CLAY EC. 2011. Evaluation and management of orthostatic hypotension. Am Fam Physician 84: 27-536.
• LAVIE CJ, LADDU D, ARENA R, ORTEGA FB, ALPERT MA & KUSHNER RF. 2018. Healthy Weight and Obesity Prevention: JACC Health Promotion Series. J Am Coll Cardiol 72: 1506-1531. https://doi.org/10.1016/j.jacc.2018.08.1037.
  » https://doi.org/10.1016/j.jacc.2018.08.1037
• LI L ET AL. 2019. Cardiovascular disease outcomes in metabolically healthy obesity in communities of Beijing cohort study. Int J Clin Pract 73: e13279. https://doi.org/10.1111/ijcp.13279.
  » https://doi.org/10.1111/ijcp.13279
• LIU C ET AL. 2019. The Prevalence of Metabolically Healthy and Unhealthy Obesity according to Different Criteria. Obes Facts 12: 78-90. https://doi.org/10.1159/000495852.
  » https://doi.org/10.1159/000495852
• MA Q ET AL. 2019. Prevalence and risk factors of prolonged corrected QT interval in general Chinese population. BMC Cardiovasc Disord 19: 276. https://doi.org/10.1186/s12872-019-1244-7.
  » https://doi.org/10.1186/s12872-019-1244-7
• MADDEN AM & SMITH S. 2016. Body composition and morphological assessment of nutritional status in adults: a review of anthropometric variables. J Hum Nutr Diet 29: 7-25. https://doi.org/10.1111/jhn.12278.
  » https://doi.org/10.1111/jhn.12278
• MAGKAS N, TSIOUFIS C, THOMOPOULOS C, DILAVERIS P, GEORGIOPOULOS G, DOUMAS M, PAPADOPOULOS D & TOUSOULIS D. 2019. Orthostatic hypertension: From pathophysiology to clinical applications and therapeutic considerations. J Clin Hypertens 21: 426-433. https://doi.org/10.1111/jch.13491.
  » https://doi.org/10.1111/jch.13491
• MANRIQUE-ACEVEDO C, CHINNAKOTLA B, PADILLA J, MARTINEZ-LEMUS LA & GOZAL D. 2020. Obesity and cardiovascular disease in women. Int J Obes 44: 1210-1226. https://doi.org/10.1038/s41366-020-0548-0.
  » https://doi.org/10.1038/s41366-020-0548-0
• MANU P, IONESCU-TIRGOVISTE C, TSANG J, NAPOLITANO BA, LESSER ML & CORRELL CU. 2012. Dysmetabolic signals in “metabolically healthy” obesity. Obes Res Clin Pract 6: e9-e20. https://doi.org/https://doi.org/10.1016/j.orcp.2011.04.003.
  » https://doi.org/10.1016/j.orcp.2011.04.003
• MATSUDO S, ARAÚJO T, MATSUDO V, ANDRADE D, ANDRADE E, OLIVEIRA LC & BRAGGION G. 2001. Questionário internacional de atividade física (IPAQ): estudo de validade e reprodutibilidade no Brasil. Rev Bras Ativ Fís Saúde 6: 5-18.
• MONGRAW-CHAFFIN M ET AL. 2018. Metabolically Healthy Obesity, Transition to Metabolic Syndrome, and Cardiovascular Risk. J Am Coll Cardiol 71: 1857-1865. https://doi.org/10.1016/j.jacc.2018.02.055.
  » https://doi.org/10.1016/j.jacc.2018.02.055
• NCD-RISC. 2016. Trends in adult body-mass index in 200 countries from 1975 to 2014: a pooled analysis of 1698 population-based measurement studies with 19·2 million participants. Lancet 387: 1377-1396. https://doi.org/10.1016/S0140-6736(16)30054-X.
  » https://doi.org/10.1016/S0140-6736(16)30054-X
• NIBOUCHE-HATTAB WN, LANASRI N, ZERAOULIA F, CHIBANE A & BIAD A. 2017. Orthostatic hypertension in normotensive type 2 diabetics: What characteristics? Ann Cardiol Angeiol 66: 159-164. https://doi.org/10.1016/j.ancard.2017.04.003.
  » https://doi.org/10.1016/j.ancard.2017.04.003
• PAFILI K, TRYPSIANIS G, PAPAZOGLOU D, MALTEZOS E & PAPANAS N. 2015. Simplified Diagnosis of Cardiovascular Autonomic Neuropathy in Type 2 Diabetes Using Ewing’s Battery. Rev Diabet Stud 12: 213-219. https://doi.org/10.1900/RDS.2015.12.213.
  » https://doi.org/10.1900/RDS.2015.12.213
• RASTOVIĆ M, SRDIĆ-GALIĆ B, BARAK O, STOKIĆ E & POLOVINA S. 2019. Aging, Heart Rate Variability and Metabolic Impact of Obesity. Acta Clin Croat 58: 430-438. https://doi.org/10.20471/acc.2019.58.03.05.
  » https://doi.org/10.20471/acc.2019.58.03.05
• RASTOVIĆ M, SRDIĆ GALIĆ B, BARAK O, STOKIĆ E & VASILJEV R. 2016. Heart Rate Variability In Metabolically Healthy And Metabolically Unhealthy Obese Premenopausal Women. Acta Endocrinol 12: 35-42. https://doi.org/10.4183/aeb.2016.35.
  » https://doi.org/10.4183/aeb.2016.35
• REYES DEL PASO GA, LANGEWITZ W, MULDER LJ, VAN ROON A & DUSCHEK S. 2013. The utility of low frequency heart rate variability as an index of sympathetic cardiac tone: a review with emphasis on a reanalysis of previous studies. Psychophysiology 50: 477-487. https://doi.org/10.1111/psyp.12027.
  » https://doi.org/10.1111/psyp.12027
• ROBERTSON D. 2011. Orthostatic Hypertension. Hypertension 57: 158-159. https://doi.org/10.1161/HYPERTENSIONAHA.110.163485.
  » https://doi.org/10.1161/HYPERTENSIONAHA.110.163485
• ROLIM LCDSP, SÁ JRD, CHACRA AR & DIB SA. 2008. Diabetic Cardiovascular Autonomic Neuropathy: Risk Factors, Clinical Impact and Early Diagnosis. Arq Bras Cardiol 90: e24-e32.
• SAMOCHA-BONET D ET AL. 2014. Metabolically healthy and unhealthy obese – the 2013 Stock Conference report. Obes Rev 15: 697-708. https://doi.org/10.1111/obr.12199.
  » https://doi.org/10.1111/obr.12199
• SHAFFER F & GINSBERG JP. 2017. An Overview of Heart Rate Variability Metrics and Norms. Front Public Health 5: 258. https://doi.org/10.3389/fpubh.2017.00258.
  » https://doi.org/10.3389/fpubh.2017.00258
• SHAW BH ET AL. 2017. Optimal diagnostic thresholds for diagnosis of orthostatic hypotension with a ‘sit-to-stand test’. J Hypert 35: 1019-1025. https://doi.org/10.1097/HJH.0000000000001265.
  » https://doi.org/10.1097/HJH.0000000000001265
• SIMS EA. 1982. Characterization of the syndromes of obesity, diabetes mellitus and obesity. Baltimore, London: Williams & Wilkins, p. 8.
• STRAUS SM ET AL. 2006. Prolonged QTc interval and risk of sudden cardiac death in a population of older adults. J Am Coll Cardiol 47: 362-367. https://doi.org/10.1016/j.jacc.2005.08.067.
  » https://doi.org/10.1016/j.jacc.2005.08.067
• TASK FORCE OF THE EUROPEAN SOCIETY OF CARDIOLOGY & THE NORTH AMERICAN SOCIETY OF PACING AND ELECTROPHYSIOLOGY. 1996. Heart rate variability: standards of measurement, physiological interpretation and clinical use. Circulation 93(5): 1043-1065. https://doi.org/10.1161/01.CIR.93.5.1043.
  » https://doi.org/10.1161/01.CIR.93.5.1043
• USUI H & NISHIDA Y. 2017. The very low-frequency band of heart rate variability represents the slow recovery component after a mental stress task. PLoS ONE 12: e0182611. https://doi.org/10.1371/journal.pone.0182611.
  » https://doi.org/10.1371/journal.pone.0182611
• VAN NOORD C, EIJGELSHEIM M & STRICKER BHC. 2010. Drug- and non-drug-associated QT interval prolongation. Br J Clin Pharmacol 70: 16-23. https://doi.org/10.1111/j.1365-2125.2010.03660.x.
  » https://doi.org/10.1111/j.1365-2125.2010.03660.x
• VANDERLEI LCM, PASTRE CM, FREITAS JÚNIOR IF & GODOY MFD. 2010. Analysis of cardiac autonomic modulation in obese and eutrophic children. Clinics 65: 789-792.
• VELHO S, PACCAUD F, WAEBER G, VOLLENWEIDER P & MARQUES-VIDAL P. 2010. Metabolically healthy obesity: different prevalences using different criteria. Eur J Clin Nutr 64: 1043-1051. https://doi.org/10.1038/ejcn.2010.114.
  » https://doi.org/10.1038/ejcn.2010.114
• VOSS A, SCHROEDER R, HEITMANN A, PETERS A & PERZ S. 2015. Short-term heart rate variability — influence of gender and age in healthy subjects. PLoS ONE 10(3): e0118308. https://doi.org/10.1371/journal.pone.0118308.
  » https://doi.org/10.1371/journal.pone.0118308
• WEISS A, CHAGNAC A, BELOOSESKY Y, WEINSTEIN T, GRINBLAT J & GROSSMAN E. 2004. Orthostatic hypotension in the elderly: are the diagnostic criteria adequate? J Hum Hypertens 18: 301-305. https://doi.org/10.1038/sj.jhh.1001668.
  » https://doi.org/10.1038/sj.jhh.1001668
• WHO. 2022. World Obesity Day 2022 – Accelerating action to stop obesity. Available at: https://www.who.int/news
  » https://www.who.int/news
• WILDMAN RP, MUNTNER P, REYNOLDS K, MCGINN AP, RAJPATHAK S, WYLIE-ROSETT J & SOWERS MR. 2008. The Obese Without Cardiometabolic Risk Factor Clustering and the Normal Weight With Cardiometabolic Risk Factor Clustering: Prevalence and Correlates of 2 Phenotypes Among the US Population (NHANES 1999-2004). Arch of Intern Med 168: 1617-1624. https://doi.org/10.1001/archinte.168.15.1617.
  » https://doi.org/10.1001/archinte.168.15.1617
• WILLIAMS SM, ELEFTHERIADOU A, ALAM U, CUTHBERTSON DJ & WILDING JPH. 2019. Cardiac Autonomic Neuropathy in Obesity, the Metabolic Syndrome and Prediabetes: A Narrative Review. Diabetes Ther 10: 1995-2021. https://doi.org/10.1007/s13300-019-00693-0.
  » https://doi.org/10.1007/s13300-019-00693-0
• YADAV RL, YADAV PK, YADAV LK, AGRAWAL K, SAH SK & ISLAM MN. 2017. Association between obesity and heart rate variability indices: an intuition toward cardiac autonomic alteration – a risk of CVD. Diabetes Metab Syndr Obes 10: 57-64. https://doi.org/10.2147/DMSO.S123935.
  » https://doi.org/10.2147/DMSO.S123935
• YAKINCI C, MUNGEN B, KARABIBER H, TAYFUN M & EVEREKLIOGLU C. 2000. Autonomic nervous system functions in obese children. Brain Dev 22: 151-153. https://doi.org/10.1016/s0387-7604(00)00094-2.
  » https://doi.org/10.1016/s0387-7604(00)00094-2
• YU TY & LEE MK. 2021. Autonomic dysfunction, diabetes and metabolic syndrome. J Diabetes Investig 12: 2108-2111. https://doi.org/10.1111/jdi.13691.
  » https://doi.org/10.1111/jdi.13691
• ZYGMUNT A & STANCZYK J. 2010. Methods of evaluation of autonomic nervous system function. Arch Med Sci 6: 11-18. https://doi.org/10.5114/aoms.2010.13500.
  » https://doi.org/10.5114/aoms.2010.13500

Edited by

• Handling editor
  Joana Gaspar

Data availability

All the raw data will be made available in the link https://osf.io/kuxag/ after acceptation of the manuscript.

Publication Dates

• Publication in this collection
  24 Aug 2026
• Date of issue
  2026

History

• Received
  01 July 2025
• Accepted
  05 Feb 2026