**1. Introduction**

Cardiovascular diseases (CVDS) mainly heart disease and stroke are the leading causes of death globaly [1]. Obesity is a major risk factor for CVDS such as coronary artery disease. However, overweight/obesity are defined as abnormal or excessive fat accumulation measured by the body mass index (BMI), but it may not correspond to the same degree of fatness and metabolic health in different individuals [2]. Thus, accurate estimation of the body composition (BC) as well as body fat distribution are relevant from a public health perspective [3]. Nevertheless, how can the true unhealthy BC and risk be measured with regard to simple baseline anthropometric measurements? In epidemiology, as in real life, not everything that seems accurate at first glance is true in reality. In medical research, false appearances and biases also occur, which can mean that valuable conclusions may turn out to be worthless. Indeed, bias in research occurs when systematic error is introduced into sampling or testing by selecting or encouraging one outcome or answer over others. Therefore, a thorough understanding about biases, and how it affects study results is essential for medical research because association of anthropometrics does not always equate to causation regarding incidents of myocardial infarction (MI) or CVD. Interestingly, this association may present effects of bias rather than reflecting the true putative risk may be responsible for all or much of the epidemiological causality. In non-randomised study designs, baseline differences in the high BC of risk or in the measured risk when comparing between healthy population and MI/CVD cases may introduce systematic bias in results. Similarly, a different BC between groups with similar baseline confounding variables may provide bias if the risk assignment does not account for the covariates that predict the receiving true risk. Thus, not all anthropometrics are optimal for risk assessment. Critical thinking that covers all potential mechanisms of bias is indispensable to prevent incorrect conclusions being drawn, which may have clinical consequences, especially when predicting MI/CVD causal risk.

Conceptually, each anthropometric provides its own biological meaning depending on the part of the BC that can be distinguished, while the notion of equality in the estimate of risk between body measurements may be respected. If not, the lack of a balanced distribution for the simple measurements between healthy and unhealthy cases will be particularly prone to the generation of false-positive results. Regarding this issue, the mathematical relation of equivalence is a key concept for specifying whether two indicators are the same with respect to a given risk. Thus, any indicator will be comparable to other or not, depending on the measured risk. Therefore, a strong association would lead us to infer or not infer a risk, given that the true nature of risk should come from the selective high risk BC instead the mere findings of the statistical association for each metric. In fact, anthropometrically-measured causal risk depends on specific bodily components; our interpretation may not be confused by the association of arithmetic indicators that suggest a supposed risk that is not verified. Thus, criteria for judgement of causal association must be respected, while also recognising that any association may be bogus, indirect o real.

## **2. Association of anthropometric measures and MI risk**

Various previous studies have recognised the association of a raised BMI with MI, as well as a higher association of abdominal obesity measures with MI [4–10]. Despite this, BMI is an important metric that has been proposed to define ideal cardiovascular health and predict CVD risk [11, 12]. However, it is only a surrogate measure of general body fatness and does not provide accurate information about the true high risk BC, unlike waist circumference (WC). Indeed, evidence is accumulating in support of WC as metric linked to visceral adipose tissue, and the only metric among

*Anthropometrics in Predicting Cardiovascular Disease Risk: Our Research Work Mathematically… DOI: http://dx.doi.org/10.5772/intechopen.105098*

other simple measurements that predict MI and cardiometabolic risk [4, 7, 9, 13–17]. However, according to the INTERHEART study and others, waist-to-hip ratio (WHR) appeared to have the best predictive value above BMI and WC [4, 18–21]. In addition, results from the UK Biobank have conferred WHR a greater excess risk for MI in women than in men [21].

On the other hand, compound metrics such as waist-to-height ratio (WHtR), whole-body fat percentage (%BF), conicity index, and adiposity measured by technological methods could be better indicators than WC alone to predict cardiovascular events and mortality, even taking consideration of sex differences [5, 14, 20, 22–27]. Furthermore, WHtR and %BF have demonstrated a high level of discrimination in the relationship with a unhealthy BC. WHtR has been more strongly correlated with %BF and adiposity variables in men than it is with WC [24, 27]. WHtR and %BF appear to be strengthened as an anthropometrically valid assessment of biological risk. Thereby, WC and height, and skin folds to a lesser extent, could be taken as basic measurements for evaluating cardiometabolic and MI risk, including cardiovascular mortality, in their relationships with abdominal and relative adiposity [12–16, 20–30]. Complementary, moderate-high endomorphy and high thickness of skinfolds, especially subscapular, have been significantly associated with MI in men [10, 24, 27, 31, 32]. Moreover, patients of both sexes assessed by computed tomography have presented better MI risk prediction as visceral adiposity increases and abdominal subcutaneous area decreases [16, 22].

## **3. What is new about anthropometrics associated with MI**

While overweight/obesity as BMI-measured, enlarged WC, WHR risk cut-off of <1, and WHtR cut-off of ≥0.5 have been verified as baseline characteristics for the association of anthropometrics and MI/CVD worldwide, even accounting for differences in strength of association and by sex [4–10, 12–19, 21–24, 27, 32–37]. Similarly, mathematical inequality between the mean simple body measurements as well as non-equivalent relation in the ratios, ratios of ratios and risk cut-offs may also be implicated (**Table 1**). Thus, data from thousands of MI/CVD cases are collated in **Table 1**, where new anthopometrics have been included as mere mathematical expressions derived from original data, demonstrating the inequality and non-equivalence relations between the corresponding mean simple measurements. After associating anthropometrics and MI/CVD risk, since mathematical inequalities between measurements may be demonstrated in any study population, perspective for epidemiological causality should be shifted accordingly. From evidence reflected in **Table 1**, neither WHR risk cutoff <1 (the mean hip circumference (HC) > WC) nor WC risk cut-off (the mean WC > height/2) will adequately describe the risk, because true risk only occurs at the volume measurement WHtR risk cut-off >0.5, where inequality between WC and height (or height/2) matters too. This is because WHtR mathematically represents a volume function with two independent factors: WC and height. These two measurements are also decisive for estimating %BF [23, 24, 27, 36, 37, 45]. In this sense, mathematical and anthropometric observations in our research work have explained the selection bias for WHR with respect to WC and WHtR and, therefore, have revealed that the risk comparison between healthy and unhealthy cases was not the same [23, 24, 27, 36, 37].

Due to anthropometrically-estimated %BF and mesomorphy presenting a high magnitude of association in MI for men [24, 27, 31], there are still uncertainties


*BMI indicates body mass index; CVD, cardiovascular disease; HC, hip circumference; Ht, body height; HHt/2R, hipto-height/2 ratio; HtHR, height-to-hip ratio; WC, waist circumference; MI, myocardial infarction; WHR, waist-to-hip ratio; WHtR, waist-to-height ratio; WHt/2R, waist-to-height/2 ratio. \*Regardless of risk cutoff values significant inequality between the mean values of the referenced simple measurements and a non-equivalent relation in the ratios is always found. \*\* Measures of association such as odds ratios, hazard ratios, Receiver Operating Characteristic curves or other statistical models in all studies were used as appropriate. (−): Null or not association. a Ethnically-specific risk cutoffs (either in numerical or in undefined values) are taken into account when reflecting inequality between the simple measurements, and therefore non-equivalent risk in the ratios, ratios of ratios and risk cutoffs. b Mathematical inequality between the simple measurements and non-equivalence relations are extracted or extrapoled from the differences between the mean (standard deviation) or median values described in thousands of participants in most studies worldwide. Table was elaborated by the author. From the scientific evidence, new metrics were included.*

#### **Table 1.**

*Defined and undefined risk cut-off points for the association of anthropometrics and MI/CVD. Imbalance between the mean values of the simple body measurements (in parentheses) where appropriate. Risk cutoff values and mathematical inequality between the corresponding simple measurements and ratios where appropriate too [4–10, 14–30, 32–35, 38–44].*

regarding the association between BMI and WHR and their relationships with the true high risk BC. Conceptually, the true risk factor regarding BC derives from %BF, fundamentally the part linked to intra-abdominal fat depots functioning as a neuroendocrine organ that influence CVD risk [46, 47]. On the other hand, mesomorphy represents relative muscularity, but association with MI is artificial and does not equate causation [10, 24, 31, 48]. Thus, seeing as BMI and WHR are anthropometrically linked to musculoskeletal component, and are more weakly correlated with %BF than other metrics, they have presented an information bias and associated a spurious risk for MI in men [23, 24, 27]. Indeed, it is important to understand the discrepancy observed between the strongest association for WHR, and their worst correlations with measures of general and central adiposity in both sexes [4, 17–19, 21, 23, 27]. The discrepancy between the strength of association for WHR and a lower anthropometric coherence as well as the unbalanced distributions for WC and HC between healthy and cases in both sexes, suggest that there where errors regarding the true risk association. Consequently, a systematic error would be introduced regarding the true risk assignment for WHR and BMI, if, when partially capturing a dimension of spurious risk their data were slanted in an artificial direction towards site of cases. In

*Anthropometrics in Predicting Cardiovascular Disease Risk: Our Research Work Mathematically… DOI: http://dx.doi.org/10.5772/intechopen.105098*

contrast, a raised WHtR and %BF have demonstrated anthropometric coherence and balanced distribution for the concrete values of volume by unit of height and body fatness for justifying risk excess. This anthropometric profile could help explain the abundance of MI among individuals with raised visceral fat, irrespective of BMI, HC or mesomorphy rating [10, 23, 24, 27, 31, 48].
