Un profil de risque cardiométabolique détérioré chez les individus avec une réponse tensionnelle excessive à l’effort sous-maximal
L’article composant ce chapitre est intitulé
«Deteriorated cardiometabolic risk profile in individuals with excessive blood pressure
response to submaximal exercise»
40 Résumé français
Titre de l’article : Un profil de risque cardiométabolique détérioré chez les individus
avec une réponse tensionnelle excessive à l’effort sous-maximal
Mise en contexte : L’identification précoce d’individus présentant un risque
cardiométabolique augmenté est une étape essentielle vers de meilleures stratégies préventives. Une réponse tensionnelle excessive à l’effort maximal a été associée à plusieurs complications de la santé.
Objectif : Documenter comment la tension artérielle à l’effort sous-maximal peut aider à
l’identification d’individus avec un profil de risque cardiométabolique détérioré.
Méthodes : Les participants (n = 3913) ont complété une évaluation cardiométabolique
exhaustive (tension artérielle de repos, circonférence de taille, profil lipidique et hémoglobine glyquée en plus d’une épreuve d’effort sous-maximal sur tapis roulant incluant un palier standardisé [3,5 mph et 2 % d’inclinaison]). Les critères d’exclusion comprenaient un historique ou des symptômes de la maladie cardiovasculaire, les femmes enceintes, l’utilisation de médication altérant la fréquence cardiaque ainsi que les blessures au bas du corps. Les participants étaient divisés en deux groupes selon la tension artérielle obtenue lors du palier standardisé (Réponse normale ou Réponse excessive : tension systolique ≥80e percentile ou tension diastolique ≥90 mmHg). Les sujets ont également été assignés aux cinq sous-groupes de tension artérielle de repos proposés par les recommandations actuelles. Les porteurs de la taille hypertriglycéridémiante ont aussi été identifiés.
Résultats : Le groupe Réponse excessive présentait une circonférence de taille
augmentée, un niveau de triglycérides augmenté ainsi qu’une condition cardiorespiratoire diminuée (P<0.01) et ce, même chez les individus normotendus (P≤0.05) au repos. Une plus grande prévalence de porteurs de la taille hypertriglycéridémiante était aussi observée pour le groupe Réponse excessive à l’intérieur de la majorité des catégories de tension artérielle de repos.
Interprétation : Cette étude suggère donc qu’une réponse tensionnelle excessive à
41
cardiométabolique qui va au-delà de ce qui est possible de prédire avec la tension artérielle de repos seulement. Par conséquent, la tension artérielle à l’effort sous- maximal représente un outil de dépistage additionnel pouvant mener à une meilleure identification d’individus à risque, pour qui des interventions préventives agressives en lien aux habitudes de vie sont nécessaires.
42
DETERIORATED CARDIOMETABOLIC RISK PROFILE IN INDIVIDUALS WITH EXCESSIVE BLOOD PRESSURE RESPONSE TO SUBMAXIMAL EXERCISE CE Côté, B.Sc.1,2, C Rhéaume, M.D., Ph.D.1,3, P Poirier, M.D., Ph.D.1,4, JP Després, Ph.D.1,2,
N Alméras, Ph.D.1,2
1Institut universitaire de cardiologie et de pneumologie de Québec – Université Laval,
Québec, QC, Canada
2Department of Kinesiology, Faculty of Medicine, Université Laval, Québec, QC,
Canada
3Faculty of Medicine, Family Medicine and Emergency Medicine, Université Laval, QC,
Canada
4Faculty of Pharmacy, Université Laval, QC, Canada
Address of correspondence: Natalie Alméras, Ph.D.
Institut universitaire de cardiologie et de pneumologie de Québec – Université Laval 2725 chemin Sainte-Foy, A-2085, Québec, QC, G1V4G5, Canada
Tel.:+1-418-656-8711 ext. 3600 Fax:+1-418-656-4527 E-mail: natalie.almeras@criucpq.ulaval.ca
Running title: Exercise blood pressure as an early risk factor
Keywords: resting blood pressure; exercise blood pressure; exercise testing;
cardiometabolic risk profile; abdominal obesity; primary prevention; cardiorespiratory fitness.
Disclosure: All authors declare that they have no conflict of interest. The Grand Défi
Entreprise Inc. was not involved in the analysis and interpretation of the data, nor in the writing of this paper.
43 Abstract
BACKGROUND: Early identification of individuals at increased cardiometabolic risk is an essential step to improve primary preventive interventions. Excessive maximal exercise blood pressure has been associated with several adverse outcomes. We examined how submaximal exercise blood pressure could help to identify individuals with a deteriorated cardiometabolic risk profile. METHODS: Data from an observational study of 3,913 participants from a convenience sample were used. Subjects included in the analyses have completed a comprehensive cardiometabolic health assessment (resting blood pressure; waist circumference; lipid profile; HbA1c submaximal treadmill exercise test including a valid standardized stage [3.5 mph and 2% slope] with blood pressure measurements. Participants were classified based on blood pressure response at the standardized exercise stage (Normal or Excessive Response). Excessive response was defined as systolic pressure ≥ 80th percentile or diastolic pressure ≥90 mmHg. Subjects were also classified into five resting blood pressure subgroups according to current guidelines. Subjects with the hypertriglyceridemic waist phenotype were also identified. RESULTS: The Excessive Response group had higher waist circumference, higher triglyceride concentrations, and lower fitness levels (P≤0.01) than the Normal Response group, even among normotensive individuals (P≤0.05) at rest. The Excessive Response group also showed a higher prevalence of hypertriglyceridemic waist subjects in most resting blood pressure subgroups. CONCLUSIONS: This study demonstrates that an excessive blood pressure response to a submaximal exercise is associated with a deteriorated cardiometabolic risk profile beyond resting blood pressure profile. Therefore, submaximal exercise blood pressure
44
represents a simple screening tool to better identify at-risk individuals requiring aggressive preventive lifestyle interventions.
45
1. INTRODUCTION
Hypertension (HT) affects almost one Canadian out of 41 and is a leading cause of the
global burden of disease in the world because of its well-documented association with strokes and ischemic heart diseases.2 Cardiovascular disease (CVD) risk starts to
increase at a blood pressure (BP) even lower than 130/80 mmHg3, which is the new
cut-off for HT diagnosis.4 This emphasizes the importance of early detection of elevated
BP, especially considering that it is estimated that 25 to 50% of the worldwide adult population is considered as having Elevated BP or Stage 1 HT.5 Lifestyle interventions
and exercise capacity improvements can prevent and delay the progression to HT.5 It is
therefore important to develop simple screening tools for early identification of individuals susceptible to develop HT and related comorbidities to target the ones who would benefit the most from aggressive lifestyle intervention.
It is known that excessive exercise blood pressure (EBP) is predictive of increased CVD mortality.6 It is also associated with higher CVD event rates7, new onset of HT7,
endothelial dysfunction8, greater elevation in ambulatory BP9, enhanced sympathetic
tone9, greater insulin resistance, greater waist circumference (WC)10, and a worsened
lipid profile.11 Therefore, EBP could help identify individuals at higher risk not only for
CVD, but also for metabolic disorders such as diabetes. However, most studies on EBP have been derived from maximal exercise testing. A recent meta-analysis has reported that EBP at moderate intensity was predictive of CVD event rates while maximal workload EBP was not.12 Other studies have also shown that the CVD predictive value
46
is seldom used in preventive interventions and requires additional resources and medical supervision in most settings. Submaximal exercise testing could represent a relevant alternative to identify individuals with an excessive EBP, thus contributing to assess cardiometabolic risk (CMR) profile. Based on the available literature on EBP, we hypothesized that individuals with an excessive EBP at submaximal exercise would be at higher risk of developing HT and related cardiometabolic comorbidities.
The objective of this study was to evaluate how EBP assessed at a standardized submaximal workload could help identify individuals with a deteriorated CMR profile. In addition, this study also aimed at documenting the safety of using a submaximal exercise testing protocol in the context of a CMR assessment performed at the workplace.
47 2. METHODS
The present cohort is part of the "Grand Défi Entreprise", a health and wellness program including a comprehensive cardiometabolic and cardiorespiratory health evaluation using a mobile risk assessment unit at the workplace.13, 14 Baseline data from
a convenience sample of 4,021 individuals (66.7% of men) who participated to the program between 2011 and 2017 were used. Irrespective of their health status, any volunteer employee could participate in the program. However, individuals with specific exclusion criteria for submaximal exercise treadmill test or invalid data for the standardized stage [3.5 mph and 2% slope] were excluded from the present cross- sectional analyses (Supplementary Flowchart). Therefore, data from 3,913 (2,632 men and 1,281 women) participants were included in the present analyses. All measurements were performed in a single session by a team of trained healthcare professionals. The local Institutional Review Board approved the study and participants provided their informed consent.
Medical history and lifestyle habits
Medical history (personal and family history as well as current medication) and lifestyle habits (physical activity level15 and nutritional quality16) were assessed by standardized
questionnaires.
Resting Hemodynamic Profile
Using appropriate cuff size, resting BP and heart rate (HR) were measured once on each arm after the participants had been sitting for at least 10 minutes, with an automated sphygmomanometer (SunTech247, SunTech Medical, Morrisville, NC, USA).
48
Left arm values were used for analyses since there was a robust correlation between systolic blood pressure (SBP) values of the left and right arms (r = 0.89).
Anthropometric Profile
According to standardized procedures, height, weight17 and WC18 were measured. Body
mass index (BMI) was calculated. Body fat percentage was estimated from a bioelectric impedance analysis scale (TBF-300A, Tanita, Arlington Heights, IL, USA).
Plasma Lipid Profile and Hemoglobin A1c
Non-fasting blood samples19 were collected from the forearm vein into lithium heparin
tubes for the measurement of plasma lipid and lipoprotein levels with a chemistry analyzer (Piccolo Xpress, Abaxis, Union City, CA, USA) using manufacturer recommended procedures. Glycated hemoglobin (HbA1c) levels were measured with an analyzer system (Cobas Integra 400/800, Roche, Mississauga, ON, Canada) based on theturbidimetric inhibition immunoassay (TINIA).20
Submaximal Exercise Testing
Cardiorespiratory fitness was assessed using a standardized submaximal exercise test performed on a treadmill (TMX425, Trackmaster, Newton, KS, USA). The protocol consisted of four stages of 2min15sec each: 1) warm-up stage (2.5 mph and 0% slope), 2) standardized stage (3.5 mph and 2% slope), 3) effort stage (personalized to reach 75% ± 5 bpm of age-estimated maximal HR), 4) cooldown stage (2.5 mph and 0% slope). An additional effort stage period was performed when target HR was not reached. Target intensity was established at 75% ± 5 bpm of HR for the effort stage so that participants would reach moderate exercise intensity, thus limiting the risk of adverse events associated with more vigorous intensity.21
Participants’ exercise HR was monitored using the Polar HR monitor system (Polar Electro, Kempele, Finland) and also using a 3-electrode electrocardiogram signal linked
49
to an automated sphygmomanometer with which EBP was assessed (Tango+, SunTech Medical, Morrisville, NC, USA). HR and EBP were measured at the end of every stage. Data collected at the standardized stage (HR, EBP and BP rise from rest to standardized stage) are presented as "exercise" data in Tables and Figures. Maximal oxygen consumption (VO2max) was estimated by linear extrapolation22 to age-estimated
maximal HR (220 - age)23 using ACSM’s Metabolic Equations24 and the least square
method.
Exclusion criteria for exercise treadmill test were related to exercise testing safety and validity. As there was no physician supervision during assessment, participants with symptoms or who had a recent CVD event did not perform the exercise test. Those with limiting lower-body injury were also excluded. Finally, participants with HR altering conditions (HR altering drugs and pregnancy) were excluded from analyses for validity purposes (influence on EBP and VO2max estimation).
Statistical Analyses
Data are presented as means±SD in Tables and means±SE in Figures. The Shapiro- Wilk test was used to examine the distribution of each variable and logarithmic transformations were applied to variables showing an abnormal distribution.
Subjects were first classified based on EBP measured during the standardized stage. Since no consensus has yet been reached in the scientific literature regarding the definition of excessive EBP response to submaximal exercise, the 80th percentile of SBP during the standardized stage for men and women was used as threshold values.25
The threshold for diastolic BP (DBP) was set at 90 mmHg since it is already used for the diagnosis of HT Stage 2 at rest4 and DBP is expected to remain stable or decrease in
50
Response to Exercise (ERE): SBP ≥ 80th percentile or DBP ≥ 90 mmHg, and 2- Normal Response to Exercise (NRE).
Participants were then assigned to one of the following categories according to their resting BP4: 1- Normal tension (NT): BP <120/80 mmHg, 2- Elevated tension (ET): BP
120-129/<80 mmHg, 3- HT Stage 1 (HT1): BP 130-139/80-89 mmHg, 4- HT Stage 2 (HT2): BP ≥140/90 mmHg and 5- Treated HT (Tx): using any hypotensive drug.
Abdominal obesity was determined using WC thresholds proposed by the International Diabetes Federation.27 Participants with the hypertriglyceridemic (hyperTG) waist
phenotype, a simple clinical marker of visceral obesity, were also identified (WC: M ≥ 90 cm, W ≥ 85 cm and triglyceride (TG) levels: M ≥ 2.0 mmol/L, W ≥1.5 mmol/L).28,29
A one-way ANOVA, adjusted for age, was performed to compare cardiometabolic parameters between groups. A P value ≤0.05 was considered statistically significant. All statistical analyses were performed using SAS statistical package version 9.4 (SAS Institute, Cary, NC, USA)
51 3. RESULTS
Participants were from 19 to 75 years of age (43.1±11.1 years) with BMI of 26.9±4.8 kg/m2, 40% were overweight, 22% were considered obese, 12% reported using lipid-
lowering drugs and 58% had an excessive WC. Elevated TG concentrations (≥1.7 mmol/L)27 were present in 42.5% of participants. Suboptimal HbA1c levels (≥5.7%) were
observed in 29% of participants while 3% had HbA1c values over the diabetes diagnostic criterion (≥6.5%).30 Based on resting BP, 22% of the individuals had NT, 16%
had ET, 30% had HT1, 22% had HT2, and 10% were using HT medications.
The 80th percentile values of SBP measured during the standardized submaximal stage (3.5 mph, 2% slope) were 175 mmHg for men and 164 mmHg for women (mean EBP at standardized stage, M: 158±22 mmHg, W: 145±25 mmHg). After classifying subjects based on exercise SBP and DBP thresholds, 24% of men and 21% of women were classified with an ERE. Table 1 shows that, after adjustment for age, ERE subjects presented deteriorated anthropometric, metabolic, resting hemodynamic, and cardiorespiratory fitness profiles compared to NRE individuals (P≤0.01). In sex-specific analyses, the ERE groups also had significantly higher WC, BMI, body fat percentage, HbA1c, TG levels,resting and exercise HR,resting SBP, resting DBP, and lower VO2max
(P≤0.01) (Supplementary Table).
Most differences between ERE and NRE remained significant within resting BP subgroups, even in the NT subgroup (Table 2). Resting HR was higher in ERE subjects in NT, HT2, and Tx groups (P≤0.01) compared to NRE. Resting SBP was higher for ERE subjects in HT1, HT2, Tx groups (P≤0.01), and also for NT (P≤0.05). Resting DBP
52
was higher in some ERE subgroups (P≤0.01). Exercise HR at the submaximal standardized stage was significantly higher in ERE than NRE subjects throughout all resting BP classes. As per class definition, exercise SBP and DBP were always higher in ERE subjects, but the SBP increase from rest to standardized exercise stage was also significantly higher for ERE in all resting BP subgroups.
Figure 1A shows that ERE subgroups always had higher WC than NRE within the same resting BP class (P≤0.01). Triglyceride levels (Figure 1B) were significantly higher in ERE subjects for the NT and HT2 subgroups compared to their NRE counterparts (P≤0.01). When using WC and TG level cut-offs defining hyperTG waist, 30.6% of the cohort had the hyperTG waist phenotype. The ERE group had a significantly higher proportion of hyperTG waist carriers than the NRE group (45.1 vs. 26.4%, P<0.01). Figure 1C shows a similar portrait within every resting BP category with ERE displaying a greater proportion of hyperTG waist carriers, especially in the ET, Tx (P≤0.01), and HT2 (P<0.05) subgroups. The ERE subjects showed lower estimated VO2max values
than the NRE participants in the ET, HT1, HT2, Tx subgroups (P≤0.01), such difference being also observed in the NT (P≤0.05) (Figure 2). HbA1c was significantly higher for ERE subjects in the HT2 subgroup than in NRE (5.5±0.5% vs. 5.7±0.6%, P<0.02). Exercise test safety
Of the initial 4,021 subjects, 50 participants were excluded from partaking the exercise test. Moreover, participants who had their test interrupted (n=187) were not included in the VO2max analyses (Supplementary Flowchart). Reasons reported for test interruption
were: target HR reached during the first two stages of the test, incapacity to maintain treadmill workload, participants requesting early ending, musculoskeletal pain or limitations, EBP >220 mmHg, and monitoring deficiencies (invalid HR or BP readings).
53
One test was interrupted due to chest pain, which was later classified as non-cardiac chest pain. No other adverse events were reported during the exercise test.
54
4. DISCUSSION
The main objective of this study was to evaluate the added value of using EBP measured at submaximal exercise in the identification of individuals at increased CMR. Our results show that individuals with an excessive BP response to submaximal exercise have an overall deteriorated CMR profile compared to those with a normal EBP response, even in individuals in NT or ET categories.
Waist circumference was significantly higher in ERE subjects. Similar associations have been reported for maximal EBP in men with the metabolic syndrome11 and for
submaximal EBP measured in asymptomatic subjects.10 Our findings extend these
associations to women and to normotensive individuals. Our results also show that normotensive individuals with an ERE had higher TG levels than NRE. It has been previously reported that individuals with both an elevated WC and increased TG concentrations, the so-called "hyperTG waist" phenotype, are at increased risk for coronary artery disease28, 29, diabetes and metabolic syndrome31, as it is a simple
clinical marker of visceral obesity in both men31 and women.32 In the present study, a
higher proportion of hyperTG waist subjects was observed in ERE subjects in most resting BP categories compared with NRE subjects. Moreover, the highest increase in hyperTG waist prevalence was found in NT (+77%) and ET (+91%) categories. Our results also show that ERE subjects had significantly higher HbA1c values than NRE subjects in men and women. Gaudreault et al.11 reached similar conclusions in a study
involving men with metabolic syndrome. In that study, individuals with elevated maximal EBP were characterized by insulin resistance. Furthermore, EBP measured at
55
submaximal exercise allowed the identification of ERE subjects who had significantly lower levels of cardiorespiratory fitness than NRE subjects, irrespective of their resting BP category. Finally, regression analyses showed that a greater SBP increase was associated with higher WC, TG levels, HbA1c values, and lower cardiorespiratory fitness (P<0.01) (results not shown). Thus, both the SBP increase to submaximal exercise and the absolute EBP value were associated with deteriorations in CMR and cardiorespiratory profiles.
Complication rates for maximal exercise testing are considered very low33 and should
logically be even lower for submaximal tests. However, scientific literature is relatively scarce on the safety of submaximal exercise testing. With no major complication reported during the testing of almost four thousand individuals, our study provides evidence that submaximal exercise testing is safe in an asymptomatic population using standardized procedures for test interruption, even when conducted without physician supervision.
The main limitation of this study resides in the fact that resting BP classification was based on a single measure of BP. Prevalence of white coat HT and masked HT may have led to misclassification of participants. Other factors as caffeine, tobacco, and stress affect the precision of BP measurement and could lead to an overestimation of elevated BP prevalence. However, we believe the size of our study samples reduces the impact of such bias, and that they are overruled by the advantages of using EBP instead of resting BP. Schultz and al.12 stated that submaximal effort offers a more
stable state for the evaluation of BP while being also more representative of the not- always-at-rest ambulatory BP assessment. In addition, non-fasting lipid concentrations
56
were used in the present study. Although this could be an issue for dyslipidemia diagnostic purposes, we have used these values only as simple clinical markers of visceral obesity. Moreover, Nordestgaard et al.34 reported that non-fasting and fasting
lipid profiles were comparable in the prediction of cardiovascular disease. Furthermore, several societies’ guidelines now endorse the use of non-fasting lipid profiles.34, 35 Thus,
we use non-fasting lipids as a screening tool without any diagnosis claim. On that basis,