We examined the association of body composition and body mass index (BMI) with submaximal systolic blood pressure (SSBP) among young adult men. The analysis included 211 men with BMI between 20 and 35 kg/m 2 . Total lean mass and fat mass were measured using dual x-ray absorptiometry and lean mass percentage was calculated from the total lean mass. Fat mass index (FMI) and BMI were calculated using height and weight (total fat mass and total weight, respectively) measurements. SSBP was measured at each stage of a graded exercise test. Quintiles of lean mass percentage, FMI, and BMI were created with quintile 1 the lowest and quintile 5 the highest lean mass percentage, FMI, and BMI. Compared with men in lean mass percentage quintile 1, those in quintiles 2, 3, and 4 had significantly lower SSBP, whereas there was no significant difference in SSBP between quintile 1 and 5 at 6, 8, and 10 minutes. Compared with men in FMI quintile 5, those in quintiles 2, 3, and 4 had significantly lower SSBP, whereas there was no significant difference in SSBP between quintile 1 and 5. SSBP among men in lean mass percentage quintile 5 and FMI quintile 1 were still less than lean mass percentage quintile 1 and FMI quintile 5, respectively. There were no significant differences in SSBP across BMI quintiles 1 to 4 but a significantly higher SSBP in quintile 5 compared with quintiles 1 to 4. In conclusion, there was a J-curve pattern between SSBP and components of body composition, whereas, a linear relation between SSBP and BMI.
Body composition and body mass index (BMI) are known predictors of cardiovascular disease risk factors, morbidity, and mortality. Studies investigating relations of submaximal systolic blood pressure (SSBP) with body composition and BMI have reported different trends, with one study suggesting that high BMI at baseline is associated with lower SSBP after 7 years in middle-aged (40 to 59 years) healthy men. Another study reported that decrease in body fat is associated with lower SSBP in older adults. To the best of our knowledge, there is no study examining the association of SSBP with body composition and BMI in young healthy men. Therefore, we analyzed the relation between SSBP and various indicators of weight and how this relation differs in case of BMI compared to body composition.
Methods
The study recruited 211 men aged 21 to 35 years with a BMI between 20 and 35 kg/m 2 . Participants were excluded if they had (1) a change in smoking status in the previous 6 months; (2) used medications to lose weight; (3) planned a weight loss surgery; (4) blood pressure (BP) at rest >150 mm Hg systolic and/or 90 mm Hg diastolic; (5) an ambulatory blood glucose level of >145 mg/dl; (6) been diagnosed with/or taking medications for a major chronic health condition; or (7) a history of anxiety, depression, or panic attack or taking selective serotonin inhibitors for any reason.
Anthropometric measurements were performed with the men wearing basic medical scrubs and in bare feet. Height (cm) and weight (kg) were measured to the nearest 0.1 cm and 0.1 kg, respectively, and BMI (kg/m 2 ) was calculated. Lean mass and fat mass were measured using a dual x-ray absorptiometry. Subsequently, lean mass percentage and fat mass index (FMI, kg/m 2 ) were calculated.
SSBP, submaximal heart rate, and cardiorespiratory fitness (CRF) defined as peak oxygen consumption (VO 2 ) were measured using a treadmill-based graded exercise test (GXT) administered by a trained exercise physiologist. The subject was instructed to rest their left arm on the shoulder of the staff member pretest and during each stage of GXT. They were also instructed to avoid gripping the shoulder of the staff member and getting an external support while walking or running during the GXT. After measurement of pre-exercise BP and heart rate, the GXT was administered using a modified Bruce protocol, which starts at a grade of 0% at 1.7 mph for initial 2 minutes and progresses to 5% grade at the same speed for next 2 minutes. After 4 minutes, the protocol is similar to that of the Bruce protocol. The speed and grade at time points 6, 8, and 10 minutes of GXT were 1.7 mph, 10%; 2.5 mph, 12%; and 3.4 mph, 14%, respectively. All men exercised to volitional fatigue and indicated when to stop. The test followed continued walking at 1.7 mph and 0% grade until their BP returned to near baseline level and heart rate ≤120. SSBP and submaximal heart rate were recorded at the second minute of each stage of the GXT by a trained staff member.
Systolic BP (SBP) was measured twice in the seated position by a well-trained technician using a mercury sphygmomanometer. The average of 2 readings was used for the representative examination value. A third measurement was taken if the difference in SBP at rest was >10 mm Hg. In addition, alcohol consumption was determined using multiple, telephone-administered, 24-hour dietary recall interviews, and smoking history was self-reported through a personal medical history questionnaire.
Quintiles of lean mass percentage, FMI, and BMI were established with quintile 1 representing the groups of men with the lowest lean mass percentage, FMI, and BMI, whereas quintile 5 represented those with the highest lean mass percentage, FMI, and BMI. One way analysis of variance was used to determine the descriptive means across all the quintiles of lean mass percentage, FMI, and BMI. The chi-square test was used to determine differences in ethnicity and smoking status in quintiles. Multivariate and univariate analysis including potential covariates including age, race, SBP at rest, alcohol intake, smoking history, and CRF were done to determine the association between SSBP at 6, 8 and 10 minutes of GXT and quintiles of lean mass percentage, FMI, and BMI. All analyses were done using SPSS (version 22) with p <0.05 for statistical significance.
A small amount of data (9%) were excluded from the current analysis because of missing SSBP or covariates. The missing SSBP data were mainly due to failure in the measurement of SSBP during the GXT when the staff member was unable to hear the Korotkoff sounds accurately during BP measurement. Nevertheless, the characteristics of men who were excluded had no significant difference (p ≥0.05) compared with those in the analysis.
Results
Table 1 lists the descriptive characteristics of men across the quintiles of lean mass percentage, FMI, and BMI. BMI quintiles 1 and 2 represent normal BMI range, quintiles 3 and 4 overweight, and quintile 5 obese. Categorizing the mean VO 2 peak according to the American College of Sports Medicine, Guidelines for Exercise Testing and Prescription , ninth edition (Philadelphia: Lippincott Williams & Wilkins, 2013) showed that CRF levels across lean mass percentage quintiles 2 to 5 were considered fair to excellent and quintile 1 as poor CRF level, whereas FMI and BMI quintiles 1 to 4 were considered excellent to fair, and quintile 5 represents poor CRF level.
| Lean mass % quintiles | Quintile 1 (N=42) | Quintile 2 (N=42) | Quintile 3 (N=43) | Quintile 3 (N=42) | Quintile 5 (N=42) | P value ∗ |
|---|---|---|---|---|---|---|
| Age, mean (SD), (years) | 28.6 (3.3) | 28.7 (3.8) | 27.2 (3.8) | 26.7 (3.6) | 26.2 (4.2) | 0.005 |
| Whites | 32 (76%) | 22 (52%) | 33 (77%) | 27 (64%) | 29 (69%) | 0.12 |
| Systolic blood pressure, mean (SD), (mmHg) | 127.3 (10.9) | 127.7 (11.2) | 124.0 (11.0) | 126.0 (10.6) | 128.4 (12.1) | 0.4 |
| Body mass index, mean (SD), (kg/m 2 ) | 29.9 (2.9) | 26.2 (2.9) | 24.6 (2.1) | 24.2 (2.2) | 23.6 (2.3) | <0.001 |
| Lean mass, % (SD) | 67.0 (3.1) | 73.3 (1.3) | 78.4 (1.6) | 83.1 (1.1) | 88.2 (2.1) | <0.001 |
| Fat mass index, mean (SD), (kg/m 2 ) | 9.8 (1.5) | 7.0 (0.7) | 5.4 (0.6) | 4.3 (0.5) | 3.1 (0.6) | <0.001 |
| Alcohol intake, mean (SD), (g/d) | 9.7 (29.0) | 10.7 (17.9) | 15.1 (20.0) | 14.2 (20.8) | 14.8 (26.1) | 0.74 |
| Vo2 peak mean (SD), (ml/kg/min) | 35.7 (5.9) | 40.3 (6.4) | 45.5 (6.3) | 47.5 (6.2) | 50.3 (7.3) | <0.001 |
| Ever smoked | 13 (31%) | 12 (29%) | 14 (33%) | 10 (24%) | 11 (26%) | 0.9 |
| Fat mass index quintiles | (N=41) | (N=43) | (N=42) | (N=43) | (N=42) | |
|---|---|---|---|---|---|---|
| Age, mean (SD), (years) | 26.1 (4.0) | 26.9 (3.8) | 26.9 (3.6) | 29.0 (3.9) | 28.6 (3.3) | 0.001 |
| Whites | 30 (71%) | 30 (70%) | 27 (64%) | 26 (61%) | 31 (74%) | 0.44 |
| Systolic blood pressure, mean (SD), (mmHg) | 128.0 (12.2) | 125.9 (11.0) | 124.6 (10.9) | 126.3 (10.4) | 128.8 (11.3) | 0.44 |
| Body mass index, mean (SD), (kg/m 2 ) | 23.5 (2.6) | 23.8 (1.9) | 24.7 (2.1) | 26.1 (2.3) | 30.6 (2.2) | <0.001 |
| Fat mass index, mean (SD), (kg/m 2 ) | 2.9 (0.6) | 4.3 (0.4) | 5.4 (0.3) | 7.0 (0.5) | 9.9 (1.4) | <0.001 |
| Lean mass, % (SD) | 88.2 (2.3) | 82.9 (1.9) | 78.6 (2.4) | 73.2 (2.2) | 67.5 (3.6) | <0.001 |
| Alcohol intake, mean (SD), (g/d) | 14.7 (26.4) | 16.1 (20.9) | 14.1 (20.1) | 9.4 (16.3) | 11.1 (30.1) | 0.66 |
| Vo2 peak mean (SD), (ml/kg/min) | 50.5 (7.7) | 48.1 (6.3) | 44.3 (5.5) | 40.8 (6.7) | 35.7 (5.4) | <0.001 |
| Ever smoked | 10 (24%) | 10 (23%) | 16 (38%) | 10 (23%) | 14 (33%) | 0.4 |
| Body mass index quintiles | (N=42) | (N=42) | (N=40) | (N=45) | (N=42) | |
|---|---|---|---|---|---|---|
| Age, mean (SD), (years) | 25.5 (3.2) | 27.0 (3.5) | 27.9 (4.1) | 28.2 (4.3) | 28.8 (3.3) | <0.001 |
| White | 26 (62%) | 32 (76%) | 24 (60%) | 28 (62%) | 34 (79%) | 0.2 |
| Systolic blood pressure, mean (SD), mmHg | 123.4 (12.6) | 124.7 (10.4) | 126.5 (10.5) | 127.2 (10.3) | 131.5 (10.6) | 0.009 |
| Lean mass, % (SD) | 83.3 (5.0) | 81.1 (6.8) | 78.5 (5.6) | 78.0 (6.3) | 69.2 (6.1) | <0.001 |
| Body mass index, mean (SD) | 21.4 (1.0) | 23.5 (0.5) | 24.9 (0.6) | 26.6 (0.8) | 30.7 (1.8) | <0.001 |
| Fat mass index, mean (SD) | 2.9 (0.6) | 4.3 (0.4) | 5.4 (0.3) | 7.0 (0.5) | 9.9 (1.4) | <0.001 |
| Alcohol intake, mean (SD), (g/d) | 11.2 (19.4) | 15.7 (26.0) | 13.1 (19.9) | 11.2 (17.8) | 14.3 (30.8) | 0.34 |
| Vo2 peak mean (SD), (ml/kg/min) | 47.2 (8.0) | 46.8 (8.0) | 45.0 (7.6) | 43.6 (7.5) | 36.9 (5.7) | <0.001 |
| Ever smoked | 9 (21%) | 15 (36%) | 9 (29%) | 14 (31%) | 13 (30%) | 0.54 |
A significant quadratic trend was evident between SSBP and the components of body composition (lean mass percentage and FMI), whereas a linear trend was displayed between SSBP and BMI ( Tables 2 and 3 ) in the multivariate analysis. Compared with men in lean mass percentage quintile 1, those in quintiles 2, and 3 had significantly lower SSBP (p ≤0.02), whereas there was no significant difference in SSBP between quintiles 1, 4, and 5 at 6 minutes after adjusting for age, race, SBP at rest, alcohol intake, smoking, and CRF. Compared with men in lean mass percentage quintile 1, those in quintiles 2, 3, and 4 had significantly lower SSBP (p ≤0.05), whereas there was no significant difference in SSBP between quintiles 1, and 5 at 8 and 10 minutes. Compared with men in FMI quintile 5, those in quintiles 2, 3, and 4 had significantly lower SSBP (p ≤0.05), whereas there were no significant differences in SSBP between quintiles 5 and 1. SSBP in men in lean mass percentage quintile 5 and FMI quintile 1 were still lower than lean mass percentage quintile 1 and FMI quintile 5, respectively. Table 4 lists that there was no significant difference in SSBP across quintiles 1 to 4 but a significantly higher SSBP in quintile 5 compared with quintiles 1 to 4 at 6, 8, and 10 minutes (p <0.02 for each and p = 0.07 for association between quintiles 4 and 5 at 10 minutes). Using VO 2 in absolute values instead of ml/kg/min in the covariates presented similar results: significant quadratic relations of SSBP with lean mass percentage, p ≤0.05 and FMI, p ≤0.03, whereas significant linear relation with BMI, p ≤0.02. Univariate analysis displayed a similar quadratic trend between SSBP and body composition (p ≤0.06 for lean mass percentage quintiles, p ≤0.03 for FMI quintiles), and a linear trend between SSBP and BMI with p ≤0.004. Analyzing relations of submaximal heart rate with body composition, and BMI showed inverse linear relation with lean mass percentage (p ≤0.01), whereas positive linear relations with FMI (p ≤0.005) and BMI (p ≤0.04) in an age, race, heart rate at rest, alcohol intake, smoking, and CRF-adjusted model.
| Minute | Quintile 1 (N=38) | Quintile 2 (N=39) | Quintile 3 (N=40) | Quintile 4 (N=36) | Quintile 5 (N=39) | P value ∗ |
|---|---|---|---|---|---|---|
| 6 | 158 (153 to 164) | 149 (144 to 155) | 149 (144 to 154) | 152 (147 to 158) | 151 (146 to 157) | 0.05 |
| 8 | 170 (164 to 176) | 162 (157 to 167) | 159 (154 to 165) | 161 (155 to 166) | 163 (158 to 169) | 0.01 |
| 10 | 184 (177 to 191) | 174 (167 to 180) | 172 (166 to 178) | 170 (164 to 177) | 175 (168 to 182) | 0.01 |
