The scaling, or normalization, of cardiovascular variables in patients, especially arterial blood pressure (BP) is critically important in the evaluation of cardiac pathologies, and a review of prior work points out the necessity for seeking further explanation of inter-individual cardiovascular variability in healthy adults (Dewey et al., 2008; Joyner et al., 2015). In particular, Dewey et al. (2008) describe how appropriate morphometric scaling of cardiac variables could help to reduce inter-individual variability in the “normal” adult population and thereby provide more accurate discrimination of pathological changes. They illustrate this by contrasting the futility of discriminating cardiomyopathy from training effects in the “healthy” population where height ranges from 160 (average Chinese woman in 2001) to 201 cm (NBA player for the 2006–2007 season) and corresponding weight ranges from 47 to 101 kg. Joyner et al speculate that inter-individual variability in physiological data, particularly cardiac output, peripheral resistance and muscle sympathetic nerve activity, might be lowered with the application of judicious scaling (Joyner et al., 2015). Unfortunately, the dependence of BP and its cardiac output (CO) and total peripheral resistance (TPR) components upon body size are relatively obscure in adult, human physiological literature.

Large, epidemiologic studies have established BP relationships related to age, gender and ethnicity (Whitlock et al., 2009; Wright et al., 2011; Hosseini et al., 2015). Two studies determined an essentially linear relationship between systolic blood pressure (SBP) and age, with greater values (~8 mmHg, normal men and ~32 mmHg, untreated hypertensive women) in oldest vs. youngest subjects (ages 18–83 years) (Wright et al., 2011; Hosseini et al., 2015). Although their SBP vs. age results were similar, these studies showed a discrepancy in diastolic blood pressure (DBP) across a 65 year time course. The study of predominantly western European and North American subjects demonstrated a decline of DBP starting around 38 years of age in women and around 48 years of age in men (Wright et al., 2011). In the study of Iranian subjects, DBP continued to rise between the ages of 25 and 69 (Hosseini et al., 2015). Other studies have also linked blood pressure variations between individuals to age (Lee et al., 1966; Jorde et al., 1986; Nardo et al., 1999; Wu et al., 2008; Joyner et al., 2015).

With respect to gender, DBP was higher in men than women in two large studies of predominantly western European and North American subjects (Whitlock et al., 2009; Wright et al., 2011), while in the study of Iranian subjects, there was no significant gender difference in DBP (Hosseini et al., 2015). In addition to autonomic activity, inter-individual blood pressure differences between healthy subjects whose arterial pressure lay in the normal range, has been linked to gender (Wu et al., 2008; Xu et al., 2013; Joyner et al., 2015). However, an interaction between gender and autonomic activity complicates this factor: Gender differences in the time course of the cardiovascular response to orthostatic challenge suggest that men show an earlier sympathetic vascular response than women who demonstrate a more prolonged vagal restraint of cardiac activity early in the response to orthostatic challenge (Evans et al., 2001; Reulecke et al., 2016; Sarafian and Miles-Chan, 2016).

Ethnicity also appears to play a role in blood pressure differences. In the Wright study, both SBP and DBP were higher in non-Hispanic, black subjects than in non-Hispanic white and in Hispanic subjects (Wright et al., 2011). This difference may be an indicator that ethnicity might also have played a role in the DBP differences seen in western European/north American vs. Iranian subjects (Wright et al., 2011; Hosseini et al., 2015).

With respect to interpersonal differences, body mass index (BMI) has been the variable most widely used to explore relationships to blood pressure (Nardo et al., 1999). A study of 894,576 people determined an approximately linear relationship between BMI and blood pressure with an increase of SBP of ~5 mmHg and DBP of ~4 mmHg for every 5 units (kg/m²) of BMI (Whitlock et al., 2009). Other studies have reported scaling by height (Arvedsen et al., 2012), by weight (Whyte, 1959; Miall et al., 1968; Jorde et al., 1986), and by genetic makeup (Xu et al., 2013).

In the cardiology literature, stroke volume (SV), CO, and TPR have been normalized by body surface area (BSA) and used to describe gender etc. responses to stress (Shoemaker et al., 2001). An effort has also been made in the cardiology literature to assess cardiovascular disease incidence by height (Paajanen et al., 2010).

In contrast to cardiology literature though, most physiologic studies of healthy adults do not scale cardiovascular variables, with notable exceptions (Arvedsen et al., 2012; Joyner et al., 2015; Sarafian and Miles-Chan, 2016). When used, scaling is mostly limited to body mass index (BMI), BSA and height, with few using weight alone as a scaling factor. In total, the underlying physiological basis for much of the between-subject variability in arterial blood pressure (BP) remains unexplained in both clinical and healthy populations.

From earliest studies (Hill and Barnard, 1897), controversy concerning sources of between- subject variability in blood pressure when supine and in response to assuming the upright posture have been debated. The controversy and the relative roles of passive gravitational effects and reflex adjustments have been comprehensively reviewed (Rowell, 1983). In general, the time course of orthostatic events upon standing or head up tilt begins with an initial drop in blood pressure attributable to the immediate effects of gravity, resulting in peripheral blood pooling that “starves” the heart of venous return. This drop in pressure is countered by reflex regulatory responses: decreased cardiac parasympathetic influence, followed by increased sympathetic and hormonal activities that act over differing time courses with the result that the heart and peripheral vasculature respond and return blood pressure to a steady state, near to that of resting levels (Rowell, 1983; Fessel and Robertson, 2006). This steady state that includes reflex responses, is used to characterize a person as having orthostatic hypotension (>20 mmHg decrease), orthostatic hypertension (>20 mmHg increase) or a “normal” (between 19 mmHg increase or 19 mmHg decrease) response to standing (Rowell, 1983; Fessel and Robertson, 2006; Yatsuya et al., 2011). Orthostatic hypertension has been ascribed to hyperactive compensatory reflex activity while orthostatic hypotension reflects an inadequate compensatory reflex response (Fessel and Robertson, 2006).

The intricacies of variability contributed by these factors alone and interacting with each other, may account for the relative obscurity of the relationship of blood pressure to weight and height in adults (Whyte, 1959; Arvedsen et al., 2012; Joyner et al., 2015; Sarafian and Miles-Chan, 2016), even though positive correlations between blood pressure and body size are well documented in children and adolescents (Gillum et al., 1982; Akahoshi et al., 1996; Patel et al., 2009; Fujita et al., 2010).

More important is the fact that few studies detail cardiac and peripheral vascular components of inter-individual blood pressure variability. The present study was conducted to obtain quantitative information concerning gender, height, weight, BSA and BMI contributions to blood pressure and its cardiac output and vascular resistance components from healthy, young men and women at supine rest and during the steady state response to standing. We hypothesized that significant variability in blood pressure at rest and standing arises from underlying relationships linking SV, CO and TPR with body size (height, weight, BSA and/or BMI).

Table 2 gives Pearson product moment correlations (r) for four indexes of body size (weight, height, BSA, and BMI) with diastolic, systolic and mean blood pressures, stroke volume, cardiac output and total peripheral resistance, at supine rest, at standing and change from supine, for the total group of subjects (men's and women's data combined). Also given are the respective p values for each correlation and the slope of the relationship. Heart rate is not included in this table, since there was no significant correlation of HR with any index of body size at supine, standing or change from supine. Figures and multiple stepwise regression results (r² and p-values, Table 3) present these findings individually and in graphical form.

At supine, there was no indication that DBP was related to body weight, height, BMI or BSA, Tables 2, 3. In marked contrast, data taken after standing for 7–10 min indicated a significant relationship between steady state DBP and each index of body size, Table 2. The relationship between DBP and body weight is shown at supine (in Figure 1A) and at standing (Figure 1B). For supine, there was no gender effect (Table 3) and therefore men's and women's data were pooled to give a single regression line. For standing (Figure 1B and Table 3), there were weight and gender effects, but no interaction, indicating that women (crosses, dashed line) had lower values than men (circles, solid line) but followed the same regression slope. In addition to being different from 0, slopes were also different between supine and standing (p < 0.0004). Continuous blood pressure recordings (Portapres) gave comparable results, not shown.

The magnitude of the change in DBP going from supine to standing provides an explanation for the difference in slopes between the two body positions. The magnitude of the increase in DBP evoked by standing was greater for subjects of greater body weight (Figure 2A), greater height (Figure 2B) or greater BSA (Table 2); the relationship with BMI was not significant. For each of these variables, there was no gender effect, so gender data were pooled. The slopes of DBP change (i.e., difference between supine and standing) vs. body weight and height (taken from Table 2), indicated that, upon standing, DBP increased ~2.5 mmHg for each 10 kg of weight, and 4.6 mmHg for each 10 cm of height in this group. These differences account for an ~17.2 mmHg greater increase in DBP for the heaviest vs. lightest of our subjects and an ~17.5 mmHg greater increase for the tallest vs. shortest of our subjects upon standing.

The relationships of SBP to indexes of body size differ from those of DBP: positive correlations of SBP with all indexes of body size were significant for both supine and standing, (Table 2). The regression relationships between SBP and weight are shown in Figure 3A for supine and in Figure 3B for standing. For SBP, there were significant gender effects, but no interactions (Table 3), so the data in Figure 3 show the same slope for men and women. Slopes from Table 2 indicate an ~4 mmHg SBP higher value for each 10 kg of body weight and ~5.9 mmHg for each 10 cm height, when supine and ~5.9 mmHg/10 kg weight and ~8.2 mmHg/10 cm height, when standing. The change in SBP going from supine to standing correlated with weight, when holding all other body size predictors constant, thus weight had the greatest influence on changes in SBP, Table 3.

The relationships between mean arterial blood pressure (MBP) and body size have a resemblance to both systolic and diastolic pressures (Table 2). When supine, significant correlations of MBP with weight and BMI resulted from systolic pressure contributions to mean pressure (since there were no significant correlations of DBP with any index of body size when supine). Standing MBP was positively related to all indexes of body size as were SBP and DBP while change in MBP was dominated by DBP as both correlated positively with height, weight and BSA.

Doppler measures of stroke volume (SV) for these subjects, indicated that supine values of SV, as well as changes in SV upon standing, correlated strongly with height, weight and BSA, but not with BMI, Table 2. There were significant gender and height effects, but no interactions, allowing us to state that the taller the person, the greater was their supine resting stroke volume. The relationship between SV and height is shown in Figure 4A for supine. We estimate that supine SV averaged ~11.5 ml greater, for 10 centimeters of height. Correlations of standing SV with height, weight and BSA were weaker than they were for supine and are not plotted, but correlations of indexes of body size with the change in stroke volume, were strong, Tables 2, 3. The relationship between height and change in SV upon standing is shown in Figure 4B. Again, there was no gender interaction so data are pooled. The taller the person, the greater their decline in stroke volume on standing, approximately 6.4 ml greater, for a 10 cm increase in height, Table 2 and Figure 4B.

The increase (compared to supine) in heart rate at 10 min of standing was 17.6 ± 5.6 beats per min. Neither supine heart rate, standing heart rate, or change upon standing, correlated with body weight, height, BSA or BMI. The largest correlation coefficient of HR with any of these variables was 0.15, not shown.

Cardiac output in the supine position, correlated with weight, height and BSA, but not with BMI, (Table 2). The relationship between supine CO and height was similar to, but weaker than, the relationship between SV and height and is not plotted. The change from supine to standing also correlated with weight, height and BSA, Table 2. The relationship between the change in CO and height was again similar to, but more moderate than, the SV response and is not plotted. As with SV, the taller the person, the greater their supine and standing CO, and decrease of CO upon standing.

Supine values of total peripheral resistance (TPR) correlated negatively with height (Table 2 and Figure 5A) with significant gender and height effects, but no interaction (Table 3). On standing, the correlation between height, weight and BSA reversed to a positive correlation (Table 2) with a slight, but significantly greater increase in women's, compared to men's, TPR (Figure 5B).

As one would expect, body height and weight were positively correlated, the greater a person's height, the greater was their weight, with the correlation more apparent in men than women, Figure 6. It is therefore not surprising that the variables we report correlate with weight, height and BSA though we note that correlations of most variables with BMI were not significant. Our data do indicate that blood pressure variables correlated more highly with weight while stroke volume, cardiac output and total peripheral resistance correlated more highly with height. For all cardiovascular variables, women and men did not differ in terms of their regression slopes; this group of women populated the lower end of the height and weight scales. However, the relationships we are reporting may not be the case for taller women; we did not have females in the higher weight/height ranges. This was the explanation proposed in a recent study reporting correlations between blood pressures and height that were readily apparent in their male, but not female, subjects (Arvedsen et al., 2012).

We observed significant positive correlations between body weight, height, BMI, and BSA with steady state, diastolic blood pressure when these young men and women were standing. Conversely, we found no such correlation with DBP in the supine state. Moreover, the more a person weighed, the taller s/he was, or the greater their body surface area, the greater their increase in diastolic pressure upon standing. In contrast, systolic pressure showed strong correlations with weight, height, BSA and BMI both supine and standing, but no significant correlations between postural changes in SBP with any index of body size. Most important, we found that inter-individual variabilities in cardiac output, stroke volume and vascular resistance related to body size, provide mechanistic explanations for the associated inter-subject variability in blood pressure that is different in supine and standing positions.

The present study identified a significant positive correlation between diastolic blood pressure and body size (weight, height, or BSA) in healthy young adults while standing, a relationship that was not present for DBP while supine. The magnitude of the standing-induced increase in diastolic pressure was also height/weight/BSA-dependent, thereby influencing standing diastolic BP. In addition, the magnitude of the standing-induced change in cardiac output was negatively related to body size, therefore, the positive correlation between peripheral resistance increase and body size accounted for the positive correlation between standing diastolic pressure and body size. We hypothesize that this relationship reflects a greater increase in vascular sympathetic activity in larger people when they undergo orthostatic stress.

At supine rest, our subjects also demonstrated a significant positive correlation between supine systolic blood pressure and body size. The mechanism for this relationship appeared to be through a positive correlation between supine stroke volume and height, which overpowered the negative correlation between resting peripheral resistance and height to produce a positive correlation between supine SBP and all indexes of body size.

In healthy, young men and women, systolic, but not diastolic, blood pressure at supine rest was positively correlated with all indexes of body size due to positive resting stroke volume and cardiac output correlations. Upon standing, the correlation between the increase of diastolic pressure with weight, height and BSA was attributable to the magnitude of peripheral vascular resistance increase that correlated with weight, height and BSA.

Our subjects were primarily young (24.4 ± 4.9 (SD) year) and therefore we could not reliably assess effects of age. In addition, there were too few obese subjects in the present study to characterize the relationship of variables to body weight in the obese range. Women's data lay in narrower height and weight ranges than did men's data, therefore, similarities between men's and women's results, may not hold for women whose height and/or weight lie outside these narrow ranges. Subsequently, the primary independent variables of the present study consisted of weight and height and their combination in the variable BSA, but not BMI. Four subjects participated in more than one study, but, because studies were conducted a year or more apart, and their weights were different, we decided to use both data points. All subjects wore Neoprene shorts at all stages of the study, and, even though effects of these shorts on blood pressure in response to standing or head up tilt were not large (Kostas, 2012), the study perhaps should be repeated in subjects unencumbered by the abdomen and pelvic compression these shorts impart. Protocols of the three studies were slightly different but it is unlikely that those differences contributed to variability in these subjects' postural control data. Cardiac output and manual blood pressure were measured 5–7 min apart perhaps contributing to error in the peripheral vascular resistance calculation. However, there was no significant difference in group peripheral resistance calculated this way and resistance calculated from mean blood pressure taken from continuous measures of BP during the time of cardiac output measurement.

JE: Co-I of the three original studies, principal author of the present study and co-author of three preceding manuscripts, wrote, revised and approved this manuscript and agrees to be accountable for all aspects of the work. SW: Data collection, analysis and manuscript development for one of the three original studies; extensive examination and statistical testing of data used in the present manuscript. CG: Responsible for data consolidation from previously conducted studies. Responsible for plots of consolidated data, interpretation and helping draft the first versions of this manuscript. VK: Data collection, analysis and manuscript development for one of the three original studies and replotting of data used in the current manuscript. CK: PI of original studies, co-author of present ms and three preceding manuscripts. QZ: Data collection, analysis and manuscript development for one of the three original studies. ER: Statistical analysis of data included in manuscript, preparation of that portion of manuscript. MS: Responsible for all aspects of cardiac Doppler data collection/analysis for original three studies plus manuscript preparation for all four studies. DR: Data critique, drafting and writing of current manuscript.