The costs of publication of this article were defrayed, in part, by the payment of page charges. This article must, therefore, be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Department of Human Nutrition, Foods, and Exercise, Virginia Polytechnic Institute and State University, 215 War Memorial Hall, Blacksburg, VA 24061. E-mail: email@example.com
Objective: We tested the hypothesis that with similar weight gain the increase in blood pressure (BP) would be smaller in men with higher cardiorespiratory fitness (HCRF) than in men with lower cardiorespiratory fitness (LCRF).
Research Methods and Procedures: Thirteen men (age = 23 ± 1, BMI = 24 ± 1) were overfed by ∼1000 kcal/d over ∼8 weeks to achieve a 5-kg weight gain. Resting BP and 24-hour ambulatory BP, body composition, and fat distribution were measured.
Results: Cardiorespiratory fitness (CRF) was higher in the HCRF group compared with the LCRF group (49.9 ± 1.2 vs. 38.1 ± 1.4 mL/kg per minute, p < 0.001). At baseline, body weight was similar in the HCRF and LCRF groups, whereas the HCRF group displayed lower levels of total body fat (13.0 ± 1.7 vs. 16.9 ± 1.3 kg, p = 0.049) and abdominal visceral fat (49 ± 6 vs. 80 ± 14 cm2, p = 0.032). Resting BP and 24-hour ambulatory BP were similar in the two groups at baseline. After weight gain, body weight increased ∼5 kg (p < 0.05) in both groups; the changes in body composition and regional fat distribution were similar. As hypothesized, the increases in resting systolic (1 ± 2 vs. 7 ± 2 mm Hg; p = 0.008) and diastolic (−1 ± 4 vs. 5 ± 1 mm Hg; p = 0.005) BP were smaller in the HCRF group. CRF was correlated with the increases in resting systolic (r = −0.64; p = 0.009) and diastolic BP (r = −0.80; p < 0.001). Furthermore, the relationship between CRF and BP remained significant after adjusting for the changes in the proportion of total abdominal fat gained as visceral fat.
Discussion: These findings suggest that higher levels of CRF are associated with a smaller increase in BP with weight gain, independently of changes in abdominal visceral fat.
Obesity represents one of the most serious public health issues facing the U.S. and other industrialized nations. Currently, more than 32% of the U.S. population is obese (1) and at increased risk for the development of cardiovascular diseases (2). Obesity and weight gain are particularly potent risk factors for the development of hypertension (3). Risk estimates from the Framingham Heart Study suggest that ∼75% and 65% of the cases of hypertension in men and women, respectively, are directly attributable to overweight and obesity (4). Importantly, long-duration obesity does not seem to be necessary to elevate blood pressure (BP)1 because the relationship between obesity and hypertension is evident in children (5) and exists throughout the entire non-obese range (6).
Despite the close relationship between obesity and hypertension, there is considerable interindividual variability in the BP response to weight change (7). This variability is likely due to a complex interaction of genetic and environmental factors (8, 9), and the identification of these factors is essential to understanding the etiology of obesity-hypertension. The results of epidemiological studies suggest that cardiorespiratory fitness (CRF) may be one such factor capable of modulating the BP response to weight change (10, 11). However, the experimental support for this postulate is currently lacking. In addition, to our knowledge, the potential impact of CRF on changes in BP throughout an entire 24-hour period [i.e., ambulatory BP (ABP)] with weight gain has not been studied.
Accordingly, we tested the hypothesis that the BP responses to modest weight gain would be attenuated in individuals with higher compared with lower CRF. In addition, because visceral body fat appears to be more closely related to BP than is subcutaneous fat (12, 13), we further hypothesized that the smaller increases in BP in individuals with higher CRF, if observed, would be associated with less visceral fat accumulation.
Research Methods and Procedures
Thirteen non-obese men (age = 23 ± 1, BMI = 24 ± 1) participated in the present study and were divided into either a higher CRF (HCRF) group or lower CRF (LCRF) group based on a median split of CRF of the entire group at baseline. The individual falling on the median was excluded from the group analyses (n = 6 per group) but was included for the correlation analyses (n = 13). All subjects were normotensive, free from overt cardiovascular and metabolic diseases, and not taking any medications. In addition, all of the individuals were sedentary to recreationally active and were weight stable (±2 kg) for 6 months before study entry. The nature and purpose of the study, along with potential risks and benefits, were explained before informed consent was obtained. The Virginia Tech Human Subjects Committee approved all experimental protocols.
Experimental Design and Protocol
After baseline testing, individuals were overfed ∼1000 kcal/d until a 5-kg weight gain was achieved. Excess calories were provided by an over-the-counter liquid dietary supplement (Boost Plus; Novartis, Basel, Switzerland; 34% fat, 50% carbohydrates, 16% protein). Progress was monitored by weekly body weight measurements and periodic consultations with a research dietitian (B.M.D.). Before post-testing, subjects underwent 4 weeks of weight stability at their elevated body weight to minimize the potential confounding effects of positive energy balance on the primary outcome variables.
For all testing sessions, subjects reported to the laboratory between 7 and 11 am after a 12-hour fast and after having refrained from exercise and caffeine for the previous 24 hours. Body weight was measured on a digital scale (nearest 0.1 kg). Body composition was determined by DXA (GE Lunar Prodigy Advance; GE Medical, Madison, WI) using software version 8.10e. Computed tomography scans (HiSpeed Cti; GE Medical) were performed to quantify abdominal fat distribution. Maximal oxygen consumption (Vo2max) was measured during graded treadmill exercise to exhaustion using open circuit spirometry (TrueMax 2400; ParvoMedics, Sandy, UT), and minutes of moderate-to-vigorous physical activity per week, which included both aerobic and resistance exercise, were obtained by self report. Vo2max measurements were not obtained after overfeeding in two individuals in the LCRF group. Resting BP was measured by mercury and automated sphygmomanometry after a 15-minute period of rest, with the subject in the seated position. Mercury sphygmomanometry measurements were always performed first to avoid bias. The BP measurements were repeated until within-session stability was achieved (±6 mm Hg on three sequential measurements) and were taken on at least three separate visits over a 2-week period to ensure between-session stability. ABP was measured every 15 to 30 minutes over a 24-hour period of normal daily activity using a portable device (Spacelabs 90207; Spacelabs Medical, Inc., Redmond WA) as described previously (14). Average ABP values were obtained for 24-hour, daytime (6 am to 10 pm), and nighttime (10 pm to 6 am) systolic and diastolic pressure. Urine sodium concentration was measured from a 24-hour collection using a Beckman Synchron LX20 Clinical Chemistry Analyzer (Beckman Coulter, Fullerton, CA).
Independent sample Student's t tests were used to compare subject characteristics and dependent variables at baseline. Repeated measures ANOVA was used to assess changes in subject characteristics and dependent variables in the HCRF and LCRF groups with weight gain. Differences in the magnitude of change in subject characteristics and dependent variables with weight gain in the two groups were also assessed with independent sample Student's t tests. Although normality tests were performed, their validity with small sample sizes is uncertain. Nonetheless, comparisons using Wilcoxon signed rank tests yielded similar outcomes. Relationships among variables of interest were assessed using simple and partial correlation analysis. All data are expressed as mean ± standard error. The significance level was set a priori at p < 0.05.
Subject characteristics are displayed in Table 1. The HCRF group was older than the LCRF group (p = 0.044). At baseline, body weight, BMI, and lean body mass did not differ (all p > 0.05) between the two groups, whereas both total body fat and percentage body fat were lower in the HCRF group (p = 0.049 and 0.002, respectively). Total abdominal fat and visceral abdominal fat were also lower (p = 0.040 and 0.032, respectively) in the HCRF group, whereas subcutaneous abdominal fat was similar (p > 0.05) in the two groups. As expected, Vo2max was higher (∼35%) in the HCRF group compared with the LCRF group, when expressed in absolute terms and relative to body weight (p = 0.006 and p < 0.001, respectively). Similarly, moderate-to-vigorous physical activity was higher in the HCRF group compared with the LCRF group (172 ± 51 vs. 25 ± 16 minutes/wk, p = 0.011).
Table 1. . Subject characteristics before and after weight gain
LCRF (n = 5)
HCRF (n = 5)
LCRF (n = 5)
HCRF (n = 5)
CRF, cardiorespiratory fitness; LCRF, lower CRF; HCRF, higher CRF; AVF, abdominal visceral fat; Vo2max, maximal oxygen consumption. Values are mean ± standard error (range).
The duration of overfeeding did not differ between the HCRF group and the LCRF group (42 ± 12 and 51 ± 13 days, respectively; p > 0.05) and was not related to CRF in the pooled sample. Furthermore, the duration of overfeeding was not related to the BP responses to weight gain (Table 2) and did not influence the relationship between CRF and BP. After overfeeding, body weight increased 5.0 ± 0.2 and 5.0 ± 0.2 kg (both p < 0.05) in the HCRF and LCRF groups, respectively. BMI, percent body fat, total body fat, lean body mass, total abdominal fat, and visceral and subcutaneous abdominal fat increased (all p < 0.05) with weight gain, and the magnitude of change was similar between groups. Vo2max decreased in the HCRF group when expressed relative to body weight (−2.8 ± 0.8 mL/kg per minute, p = 0.009) but not when expressed in absolute terms (0.03 ± 0.07 L/min, p > 0.05); neither expression changed in the LCRF group (Vo2max was not measured in two individuals after overfeeding). Heart rate did not change significantly with weight gain in either group (p > 0.05).
Table 2. . BP before and after weight gain
BP, blood pressure; CRF, cardiorespiratory fitness; LCRF, lower CRF; HCRF, higher CRF; ABP, ambulatory BP. Values are mean ± standard error (range).
Manual diastolic pressure and 24-hour and daytime ambulatory diastolic pressure were lower in the HCRF group compared with the LCRF group at baseline (p = 0.005–0.049) (Table 1). All other indices of resting BP and ABP were similar between groups at baseline (p > 0.05) (Table 1). After weight gain, the HCRF group experienced smaller increases in manual systolic BP (2 ± 1 vs. 7 ± 3 mm Hg, p = 0.048, Figure 1A) and in automated systolic BP (1 ± 2 vs. 7 ± 2 mm Hg, p = 0.008, Figure 1C) and diastolic BP (−1 ± 4 vs. 5 ± 1 mm Hg, p = 0.005, Figure 1D) compared with the LCRF group. Resting manual diastolic BP did not change significantly in either group (0 ± 2 vs. 2 ± 2 mm Hg, p > 0.05, Figure 1B).
The increase in 24-hour systolic BP (3 ± 1 vs. 6 ± 2 mm Hg, p = 0.081, Figure 2A) and diastolic BP (−1 ± 1 vs. 3 ± 2 mm Hg, p = 0.079, Figure 2B) with weight gain tended to be smaller in the HRCF group compared with the LCRF group. The magnitude of change in nighttime systolic (−1 ± 3 vs. 8 ± 2 mm Hg, p = 0.009, Figure 2C) and diastolic (−3 ± 3 vs. 4 ± 2 mm Hg, p = 0.029, Figure 2D) pressures was significantly smaller in the HCRF group. The change in daytime systolic pressure was similar in the HCRF and LCRF groups (5 ± 2 vs. 4 ± 3 mm Hg, p > 0.05), whereas daytime diastolic pressure remained unchanged in both groups (p > 0.05). Urinary sodium excretion was lower in the HCRF group vs. the LCRF group at baseline (111.8 ± 15.5 vs. 167.8 ± 17.4 mM, p = 0.019) but did not change in either group after weight gain (p > 0.05).
CRF was inversely correlated with the increases in resting automated systolic (r = −0.64; p = 0.009; Figure 3A) and diastolic BP (r = −0.80; p < 0.001; Figure 3B) in the pooled sample. The change in abdominal visceral fat (AVF) was not related to changes in BP, although the proportion of total abdominal fat gained as AVF was positively correlated with the changes in resting manual systolic (r = 0.50, p = 0.042) and diastolic (r = 0.56, p = 0.023) BP. The relationship between CRF and both systolic (r = −0.61, p = 0.017) and diastolic (r = −0.79, p = 0.001) BP remained significant after statistically controlling for the changes in the proportion of total abdominal fat gained as visceral fat.
The major new finding of the present study was that modest, diet-induced weight gain increases resting BP and ABP in normotensive, non-obese individuals, but the magnitude of increase is significantly smaller in individuals with higher compared with lower CRF. This protective effect of CRF appears to be independent of changes in AVF.
The results of epidemiological studies indicate that high levels of CRF or physical activity are associated with smaller increases in BP with weight gain (10, 11). The results of the present experimental study extend those findings by demonstrating that higher levels of CRF attenuate the increases in both resting BP and ABP after modest, diet-induced weight gain in normotensive, non-obese individuals. In light of recent evidence suggesting that even small elevations in BP within the normotensive range increase cardiovascular and overall mortality (15, 16), the smaller increase in BP in individuals with higher CRF in the present study may be clinically relevant (17).
The results of numerous studies have supported a close relationship between abdominal fat and BP (13, 18). In addition, CRF is associated with lower levels of visceral fat independently of total adiposity (19). However, whether visceral fat mediates the relationship between CRF and BP is less clear. The results of a recent cross-sectional study suggest that high CRF is associated with lower BP at levels of visceral fat typically observed in non-obese individuals (20). Consistent with this, our current findings suggest that CRF attenuates the increase in BP with weight gain independently of the amount of visceral fat gained. Our findings suggest that the deleterious effects of weight gain on BP may be mitigated by maintaining higher levels of CRF. However, it is unclear whether the protective effects of fitness observed in the present study would be as apparent after larger increases in total and visceral adiposity.
CRF is a complex phenotype influenced by both genetic and environmental (e.g., physical activity) factors. In this context, it is unclear whether the smaller increase in BP in individuals with higher CRF is a reflection of genetic factors or greater physical activity levels compared with the lower CRF group. Interestingly, rats artificially bred for high aerobic capacity (i.e., CRF) demonstrate lower daytime, nighttime, and 24-hour mean BP (and several other cardiovascular risk factors) compared with animals with lower aerobic capacity (21). Importantly, the more adverse cardiovascular risk factor profile was observed in 5-week-old pups before differences in body weight and visceral fat accumulation were evident. The latter observation is consistent with our current findings and the results of exercise training studies indicating that BP lowering is independent of changes in body composition (22, 23).
There are some limitations of the current study that warrant consideration. First, the sample size of our study was small. As such, inclusion of a larger number of subjects may have produced a different outcome. Second, the results of studies in rodents suggest that gender (24) and age (25) can influence the BP response to weight gain. Therefore, our results should not be extrapolated beyond the population studied. Third, our study was not designed or sufficiently powered to determine the mechanisms responsible for the smaller increase in BP in men with higher compared with lower CRF. Future studies will be necessary to determine the potential roles of the sympathetic nervous and renin-angiotensin-aldosterone systems. Finally, as mentioned above, CRF is determined by both genetic and environmental factors, but the design of the present study did not allow us to distinguish the relative importance of these factors in mediating the effect of CRF on the BP response to weight gain. A prospective study designed to investigate the influence of regular aerobic exercise on the BP response to weight gain would be necessary to address this issue.
In conclusion, the results from the current study suggest that higher levels of CRF are associated with smaller increases in resting BP and ABP after modest, diet-induced weight gain in healthy men. The beneficial effects of CRF on the BP response to weight gain seem to be independent of changes in visceral fat. Taken together, our findings may have important implications for understanding why weight gain increases cardiovascular disease risk in some individuals more than others and highlight the importance of maintaining regular physical activity levels even during periods of a positive energy balance.
This work was supported by NIH Grants HL62283 and HL67227 (to K.P.D.). We thank Emily Van Walleghen for technical assistance and the study participants for their time and cooperation.