As early as in 1977, when I arrived as a freshman at Université Laval, I had a keen interest for human body fat, its morphology, metabolism and to what extent it would be affected by regular endurance exercise. Luckily, one of the initial courses that I attended was given by a scientist who would later become a tremendous mentor for me, Professor Claude Bouchard. At that time, he was looking for research assistants to conduct a study that would explore the genetic aspects of various determinants of human health including cardiorespiratory fitness, adiposity, blood pressure and lipids, as well as the role of relevant behaviours such as diet and physical activity/exercise. To examine how all these variables were inter-related in modulating health risk was also part of the research agenda. At 18 years of age, it was for me a great privilege to be introduced to health research through a cohort study that would generate a large number of peer-reviewed publications: the Québec Family Study (QFS) [1].

Early enough in my academic training, I was struck by the interest of participating parents for their own health and also by their concerns regarding the behaviors and health of their children. Among all the variables that were measured in QFS, individual variation in body fat and its consequence on health was for me a fascinating topic. Thus, after graduating, I was grateful that my mentor accepted to develop a human adipose tissue laboratory where we would have the opportunity not only to assess body composition, but also to obtain from volunteers samples of adipose tissue from biopsies in order to study adipose tissue morphology and metabolism and its response to endurance exercise training.

During my graduate studies, we also published results of analyses conducted on adult participants of the QFS cohort that would leave me perplexed. For instance, while we reported a negative correlation between the amount of body fat and serum HDL-cholesterol levels in men (a rather expected finding from the literature already available at that time), no such relationship was observed in women [2]. Not only women had systematically higher HDL-cholesterol levels than men despite having more body fat, women with high levels of body fat showed no reduction in their serum HDL-cholesterol concentrations. Thus, their increased adiposity had little or no detrimental effect on their lipid profile compared to men [2]. At that time, underwater weighing was the technique used to assess body density, from which quantities of fat-free mass and fat mass were derived assuming different average density values of fat versus non-fat tissues [3].

Late 80’s: Imaging Methods to Study the Link Between Adiposity and Type 2 Diabetes or Cardiovascular Disease Risk − A Game Changer

In 1983, a group of Japanese investigators reported evidence that computed tomography (CT) could be used to generate images of the abdomen where one could easily distinguish adipose tissue from muscle and bone tissues on the basis of differences in tissue density (attenuation values expressed in Hounsfield units) [4]. With this technique, they provided evidence that 1- there were remarkable individual differences in the amount of visceral (VAT) versus subcutaneous adipose tissue (SAT); 2- the health consequences of such differences in VAT versus SAT were not trivial. For instance, at any body composition value or body weight, individuals with high levels of VAT appeared to be characterized by metabolic complications predictive of an increased risk of type 2 diabetes and cardiovascular disease (CVD) [5,6,7]. These fascinating early findings struck me and would determine what I would do next as a young independent investigator. Thus, after setting up my own research laboratory in 1986, I began recruiting asymptomatic men and women to document the associations between VAT measured by CT and its relationships with metabolic risk factors for type 2 diabetes and CVD.

Our early analyses conducted on our initial cohort of asymptomatic and healthy volunteers revealed that they were indeed remarkable individual differences in the amount of VAT that could not be predicted from the presence/absence of an overweight/obesity status. Indeed, some individuals who would have a diagnosis of obesity on the basis of their body mass index (BMI) would nevertheless have low levels of VAT. At the opposite, some other participants with presumably normal BMI values would show excess levels of VAT [8, 9]. Despite being matched for their amount of total body fat, a subgroup of individuals with a high accumulation of VAT would be characterized by evidence of insulin resistance, glucose intolerance, and compensatory hyperinsulinemia (even in the absence of type 2 diabetes) compared to a subgroup of participants with low levels of VAT [8, 9]. Furthermore, individuals with excess VAT would have a typical atherogenic dyslipidemic state: elevated triglyceride concentrations, increased apolipoprotein B levels, increased proportion of small LDL and HDL particles, and low HDL-cholesterol concentrations [10,11,12]. Of course, we recognize the above abnormalities as well-known features of the insulin resistance syndrome (Syndrome X initially proposed by the late Gerald Reaven) [13] or of the metabolic syndrome (a concept developed by a team of experts led by Professor Scott Grundy) [14].

At that time, we also reported that at any given level of total body fat, men would have on average about twice the amount of VAT observed in premenopausal women [15]. We then found that this phenomenon partly explained why premenopausal women are relatively protected from the deleterious effect of increasing levels of body fat on several features of their cardiometabolic risk profile: they accumulate more fat in their SAT (and therefore less in their VAT) compared to men [16].

The 90’: Metabolic Markers of Visceral Obesity Increasing CVD Risk − Evidence from the Québec Cardiovascular Study

The two principal investigators of a prospective cardiovascular epidemiology study (The Québec Cardiovascular Study [QCS]) conducted at the Québec Heart and Lung Institute (Institut universitaire de cardiologie et de pneumologie de Québec – Université Laval), Dr. Gilles Dagenais and Dr. Paul Lupien, gave to my research team the remarkable opportunity to test the hypothesis that some of the features of the insulin resistance/metabolic syndrome linked to excess VAT would be associated with an increased risk of coronary heart disease (CHD). Through a series of papers published on this cohort of middle-aged men, we reported that apolipoprotein B concentration was the feature of the lipoprotein-lipid profile most closely and independently associated with CHD risk [17]. Furthermore, the combination of increased apolipoprotein B and small LDL particles (a signature feature of the atherogenic dyslipidemia of visceral obesity, irrespective of LDL-cholesterol levels) was also associated with an increased CHD risk [18, 19].

In addition, as we had previously reported that hyperinsulinemia, as an early marker of insulin resistance among individuals without type 2 diabetes, was also a feature of visceral obesity, we did examine the association between fasting insulin levels and CHD risk in the QCS. Even among men without diabetes, we found that men in the top tertile of fasting insulin levels were at increased risk of CHD even after control for traditional CVD risk factors [20]. Furthermore, we later reported that the triad of fasting hyperinsulinemia, elevated apolipoprotein B concentration and small LDL particles was associated with a 20X increase in CHD risk, making the simultaneous presence of these features of visceral obesity an atherogenic metabolic triad [19]. As abdominal obesity is associated with a state of low chronic inflammation (as reflected by increased C-reactive protein [CRP] levels) we also reported a close relationship between waist circumference and CRP levels [21]. On that basis, we proposed that the most prevalent phenotype associated with high CRP concentrations was abdominal obesity.

Despite these advances, we already recognized at that time that it would not be realistic to propose that the above features of the atherogenic metabolic triad should be measured in primary care. Therefore, we were interested in developing simple metrics to identity individuals with features of the atherogenic metabolic triad in proximity health services. In 1994, we suggested that a simple anthropometric measurement, waist circumference, could be a useful marker of abdominal adiposity, beyond the information provided by the BMI [22]. However, as an enlarged waist circumference could be resulting from an increased accumulation of either SAT or VAT, we were keen to identify a simple clinical marker that would help us in the discrimination of visceral from subcutaneous obesity. In 2000, we reported that a commonly measured blood lipid marker, triglycerides (TG), could be helpful when combined with waist circumference for the identification of individuals likely to be characterized by visceral obesity, this adiposity phenotype being closely associated with increased circulating TG levels [23]. We proposed to name this new phenotype “hypertriglyceridemic waist”. Several prospective studies have since reported that hypertriglyceridemic waist is not only a useful marker of visceral obesity but that its presence is predictive of an increased risk of CVD [24, 25].

As excess visceral adiposity is a central feature of the insulin resistance syndrome, it is therefore of no surprise that, in the development of simple tools and criteria to diagnose its presence, the NCEP-ATP III committee had included waist and TG, in addition to HDL-cholesterol, blood pressure and glycemia as variables to diagnose a condition that they referred to as the metabolic syndrome (rather than to insulin resistance syndrome as no measurement of insulin resistance was used for its diagnosis) [14].

The new Millennium: Dysfunctional Adipose Tissue and the Concept of Cardiometabolic Risk

Since the introduction of the metabolic syndrome concept and despite the fact that PUBMED can track more than 150,000 papers on the topic, whether a clinical diagnosis of the metabolic syndrome as a distinct clinical entity exacerbates CVD risk beyond the contribution of its individual components remains debated [26, 27]. Regarding this controversy, some key issues should be highlighted: 1- irrespective of the presence or not of a multiplicative effect on CVD risk, presence of metabolic syndrome is associated with an increased CVD risk; 2- metabolic syndrome cannot be used to assess CVD risk as it does not fully capture the contribution of traditional CVD risk factors.

Because of the latter limitation of metabolic syndrome as a CVD risk calculator, we discussed in 2006 the concept of cardiometabolic risk, as an approach to incorporate features of the metabolic syndrome (mainly visceral obesity) to traditional risk factors, with the ultimate goal of improving the discrimination of overall CVD risk (cardiometabolic risk) [28]. At that time, we proposed that the so-called “residual” CVD risk not explained by traditional risk factors could be the consequence of excess visceral adiposity, which could be crudely assessed by the simultaneous presence of elevated waist and TG levels (hypertriglyceridemic waist).

Another major issue regarding the pathophysiology of insulin resistance and features of the metabolic syndrome was why people would put on SAT versus VAT. On the basis of the evidence available at that time, we proposed that excess visceral adiposity was the result of the relative inability of SAT to expand as a protective metabolic sink when facing a positive energy balance [28]. Over the last 20 years, considerable evidence has revealed that subcutaneous adiposity, particularly lower body gluteal/femoral adipose tissue, could act as a “metabolic sink”, providing protection against accumulation of body fat in diabetogenic/atherogenic depots such as VAT and unwanted accumulation of lipids in normally lean tissues such as the heart, the liver, the skeletal muscle, the kidney and the pancreas, a phenomenon referred to as “ectopic fat deposition” [28,29,30,31,32] (Fig. 1). Discussing the evidence behind this theory is beyond the scope of this brief narrative review but the concept of dysfunctional subcutaneous adiposity explaining why the resulting lipid spillover gets accumulated not only in the VAT but also in normally lean tissues has received support from many large cardiometabolic imaging cohort studies and integrative physiology studies conducted on human adipose tissue [29, 33,34,35]. Considerable amount of work is currently performed by several laboratories around the world to better understand the specific contributions of these ectopic fat depots (e.g. VAT versus liver steatosis) to various clinical outcomes.

Fig. 1

The “metabolic sink” model. Under this theory, the ability of subcutaneous adipose tissue (AT) to expand through hyperplasia (new fat cells) when facing an energy surplus protects normally lean tissues such as the liver, heart, kidney, skeletal muscle, pancreas against harmful accumulation of lipids, a phenomenon described as ectopic fat deposition. Although the concept of metabolically healthy obesity is debated, considerable evidence indicates that subcutaneous obesity is less detrimental to cardiometabolic health than visceral obesity, where relative inability of subcutaneous AT to expand (when facing a positive energy balance) leads to ectopic fat deposition, visceral AT accumulation and to the development of components of the metabolic syndrome for which insulin resistance is a key feature. BMI: body mass index, CVD: cardiovascular disease, IGT: impaired glucose tolerance, T2D: type 2 diabetes

2010: From CVD Management to Promotion of CV health − a Quiet Revolution

In 2010, a large group of experts led by Donald Lloyd-Jones met to discuss and develop the strategic objectives of the American Heart Association (AHA) until 2020 [36]. This initiative led to a remarkable breakthrough: move the focus from CVD management to the promotion of cardiovascular health. This paradigm shift would imply that, in addition to secondary and primary prevention, AHA would promote the concept of primordial prevention, that is, avoiding the development of altered CVD risk factors in the first place, a much more ambitious objective. In order to promote that concept, the committee had to agree on how to define cardiovascular health. On the basis of the evidence available at that time, showing that behaviors were important modulators of CVD risk beyond traditional CVD risk factors, they proposed 7 criteria that they would refer to as the Life’s Simple 7. In addition to being free from clinical signs of CVD, individuals in ideal cardiovascular health would have normal cholesterol, blood pressure, and glucose levels but also healthy behaviors such as not smoking, having a normal BMI (as a crude marker of being in energy balance), be physically active and have an overall healthy diet.

Cohort studies conducted using these simple metrics of cardiovascular health would then show a remarkable discrimination of CVD risk. Indeed, extremely low incidence of CVD has been reported among individuals in ideal cardiovascular health whereas a dramatic increase in incidence of CVD events has been observed with the cumulative addition of unmet criteria of cardiovascular health [37,38,39,40]. A striking and important finding was that even among individuals with normal levels of traditional biological risk factors (normal cholesterol, blood pressure and glucose levels), having unhealthy behaviors was associated with a substantial increase in CVD risk [37].

Thus, the concept of ideal cardiovascular health has revealed to what extent behaviors matter in addition to biological risk factors. Unfortunately, the prevalence of ideal cardiovascular health is extremely low (below 1%) all over the world [37, 38, 41, 42]. This could be considered as either bad or good news. However, on the positive side, such a low prevalence implies that one could substantially improve the cardiovascular health of our populations by public health measures to make healthy behaviors the “default” behaviors and by the development of clinical approaches targeting smoking cessation, healthy eating, physical activity and unhealthy adiposity (not body weight for reasons described above). Recently, AHA has also included sleep as an important health behavior and now promotes the concept of Life’s Essential 8 (Fig. 2) [43]. However, the key message behind this paradigm shift remains the same: behaviors matter.

Fig. 2

The concept of ideal cardiovascular health promoted by the American Heart Association (AHA). Ideal cardiovascular health is defined by healthy behaviors (sleep was recently added to AHA Life’s Simple 7 to make it AHA Life’s Essential 8). This concept emphasizes the notion that in order to maintain/achieve ideal cardiovascular health, behaviors matter as much as traditional biological risk factors. * Behaviors not only impact cardiovascular health directly but also affect it through their documented effects on biological risk factors. † Author’s proposed consideration for the inclusion of body mass index (BMI)-specific categories of waist circumference values considering the documented value of this approach in the prediction of morbidity/mortality risk

2010–20: Physical Activity as a Vital Sign in Clinical Practice

It is now very well documented that regular physical activity/exercise has tremendous benefits on all dimensions of health [44, 45]. Our laboratory has extensively studied the effects of regular endurance exercise on body composition, visceral adiposity and related features of cardiometabolic risk [46,47,48,49,50,51,52]. Reviewing these findings is beyond the scope of this brief narrative review. Despite evidence that regular endurance exercise (such as brisk walking, or cycling) could induce a lost of VAT (even in the absence of weight loss), maintain or even increase lean body mass, improve insulin sensitivity and glucose tolerance, and all features of the metabolic syndrome, I have personally failed to convince my own health service authorities to incorporate these notions in primary health care. Unfortunately, despite overwhelming evidence of their importance, we still do not measure behavioral features of Life’s Essential 8 in clinical practice.

A decade ago, with the help of a Foundation (Grand défi Pierre Lavoie), we were able to put in place a work health promotion program where we have transformed a bus into a mobile cardiometabolic-cardiorespiratory unit [53,54,55,56,57,58,59]. As this workplace health program was not funded by our public healthcare system, employers would pay for this initiative and offer it to their work force. The principles behind this workplace health program are simple: 1- if behaviors are important (in addition to biological risk factors), they should be assessed and targeted; 2- individual counselling based on participants’ data is also key, particularly when dealing with behaviors since they are not assessed in publicly funded primary care; 3- ongoing interactions with the research team (individual problem solving) is important; 4- group dynamics are helpful to change norms within whichever company involved in the program; and 5- follow-up measurements of behaviors and related relevant biological outcomes (including providing individual feedback) are key information to provide to participants.

On the basis of our previous work, the comprehensive set of measurements that would be offered in this workplace health promotion program would place a major focus on what we have referred to as key “lifestyle vital signs”: waist circumference (as a marker of abdominal adiposity, cardiorespiratory fitness (as the most critical determinant of life health trajectories), overall diet quality (measured by a food-based questionnaire) and level of physical activity (assessed by a standardized questionnaire).

Several papers have reported the effects of this workplace health program [53,54,55,56,57,58,59]. Here are the main findings published. 1- As reported in studies on Life’s Simple 7 or Life’s Essential 8, we found that our four “lifestyle vital signs” were powerful correlates of levels of traditional CVD risk factors [59]. In other words, among workers with healthy lifestyle vital signs, the prevalence of altered levels of CVD risk factors was very low. 2- Both overall diet quality and level of physical activity were associated with the presence/absence of hypertriglyceridemic waist