Determinants of systolic blood pressure response during exercise in overweight subjects

Abstract

Aim. Higher systolic exercise blood pressure (BP) is associated with increased cardiovascular risk in hypertension. We aimed at identifying covariates of systolic exercise BP in overweight subjects. Methods. 77 subjects with body mass index (BMI) > 27 kg/m² and without known heart disease were tested. BP was measured by sphygmomanometry before and at all exercise stages during maximal exercise capacity testing on a treadmill. High peak systolic exercise BP was defined as ≥ 200 mmHg. Results. The study population was 48 ± 10 years and included 60% women and 42% with known hypertension. Average BMI was 32.6 ± 4.8 kg/m² and clinic BP 132/82 ± 17/8 mmHg. High systolic exercise BP was found in 32%. Subjects with high systolic exercise BP had higher systolic clinic and 24-h ambulatory BP (ABP), as well as lower peak oxygen uptake, compared with subjects with normal systolic exercise BP (all p < 0.05). In multiple regression analysis known hypertension (β = 0.33), higher systolic ABP (β = 0.22) and high-density lipoprotein (HDL)-cholesterol level (β = 0.23, all p < 0.05) predicted higher systolic exercise BP independent of sex and peak oxygen uptake (multiple R² = 0.32, p < 0.001). Conclusion. Among overweight subjects, known hypertension, higher systolic ABP and HDL-cholesterol level were the most important factors predicting higher systolic exercise BP.

Key Words::

• Ambulatory blood pressure
• exercise blood pressure
• hypertension
• obesity
• overweight

Introduction

Overweight and obesity has reached pandemic proportions in the world. Once considered a problem only in high income countries, overweight and obesity are now dramatically on the rise in low- and middle-income countries, being the fifth leading cause for global deaths, and attributable to 44% of the diabetes burden and 23% of the ischemic heart disease burden (1,2). Obesity is associated with several pathophysiological mechanisms, including impaired arterial compliance, insulin resistance and increased insulin-mediated sympathetic stimulation, resulting in hypertension and type-2 diabetes and high cardiovascular disease risk (3–5).

Systolic blood pressure (BP) ≥ 200 mmHg during exercise testing (high systolic exercise BP) has been associated with increased risk of cardiovascular events and presence of subclinical cardiac target organ damage in hypertension in previous reports (6–9). High systolic exercise BP has also been associated with presence of masked hypertension and increased risk of future development of hypertension in subjects with normal clinic BP (10,11). Recently, Karavelioglu et al. (12) demonstrated increased systolic BP response in normotensive type 2 diabetic patients, while Huot et al. (13) reported elevated systolic exercise BP in individuals with increased waist circumference regardless of their insulin sensitivity or level of cardiorespiratory fitness. The present study was undertaken to further identify factors associated with higher systolic exercise BP in overweight and obese individuals.

Methods and Materials

Study population

Starting in October 2009, men and women aged 30–65 years with body mass index (BMI) > 27 kg/m² were invited by their general practitioner (GP) to participate in the ongoing FAT associated CardiOvascularR dysfunction (FATCOR) project at Haukeland University Hospital in Bergen, Norway. Exclusion criteria were previous myocardial infarction, gastrointestinal disorder, severe psychiatric illness and inability to understand Norwegian language. Out of the 180 subjects that were enrolled in the FATCOR-study by the end of 2011, the 77 subjects with complete data sets within December 2011 were included in this pilot study, the remaining subjects not being lost to follow-up but waiting to complete their examinations and therefore not included here. The FATCOR project was approved by the Regional Ethics Committee. All participants signed informed consent.

Measurements

Cardiovascular risk factors and medical history were reported by the participants on a standardized questionnaire quality assured by the GP. A general medical examination including body composition analysis (Tanita BC-420MA, Tokyo, Japan), office BP, blood and urine samples was performed at the GP center.

Office BP was measured in the sitting position after an initial rest of at least 10 min, following the European Society of Hypertension guidelines (14). BP was measured at least twice with 1-min intervals, and office BP was taken as the last measurement in individual patients, using an Omron M4 automatic sphygmomanometer (Omron Healthcare Co. Ltd., Hoofddorp, Netherlands), operating on the oscillometric principle. Pulse pressure was calculated as the difference between systolic and diastolic BP, and mean BP as diastolic BP + 1/3(pulse pressure).

Ambulatory blood pressure monitoring

Twenty-four-hour ambulatory BP (ABP) recording was performed using a non-invasive ambulatory BP monitor Diasys Integra II (Novacor, Cedex, France), set to auscultatory mode. The BP monitor was mounted with an appropriately sized cuff, and the participants instructed to relax their arm when readings were initiated. BP was measured every 20 min during daytime and every 30 min during night-time, giving an average of 78 measurements per 24 h. Daytime was defined according to diary in individual subjects. The ABP recording was only accepted when > 70% of the measurements were technically valid (14).

Exercise testing

All participants underwent a maximal exercise capacity test on treadmill following the Chronotropic Assessment Exercise Protocol (CAEP) to assess aerobic capacity and BP response (15) (Table I). The ergospirometry system used was a Schiller CS-200 (Schiller AG, Baar, Switzerland). BP measurements during exercise were carried out using a Schiller BP-200 Plus monitor, with readings initiated manually by specially trained nurses. Exercise BP was measured before initiation of exercise, at every exercise step, at termination of exercise and after 1 and 5 min of resting after completion of the test. During exercise, only systolic BP was considered. All participants were encouraged to continue exercise until exhaustion in the absence of symptoms (chest pain, breathing problems), or ischemia or ventricular tachyarrhytmias on the electrocardiogram observed by the attending cardiologist. For safety, a peripheral venous cannula was also introduced before exercise. High systolic exercise BP was defined as systolic BP ≥ 200 mmHg in line with previous publications in the field (9,11).

Table I. Excerpt from the Chronotropic Assessment Exercise Protocol.

Stage	Time (mm:ss)	Cumulative time (mm:ss)	Speed (km/h)	Grade (%)	METs (1 MET = 3.5 ml O2/kg/min)
0	0:00	0:00	0.0	0	1
1	2:00	2:00	2.2	3.0	2.7
2	2:00	4:00	2.2	4.0	3.0
3	2:00	6:00	2.7	5.0	3.6
4	2:00	8:00	3.3	6.0	4.3
5	2:00	10:00	3.8	8.0	5.3
6	2:00	12:00	4.4	10.0	6.1
7	2:00	14:00	5.5	10.0	6.4
8	2:00	16:00	6.4	10.0	6.4

Statistical analysis

Data management and statistical analysis was performed using SPSS version 19 statistical software (IBM SPSS Statistics, Chicago, IL, USA). The study population was divided into groups of patients with normal or high systolic exercise BP (peak systolic exercise BP ≥ 200 mmHg). Continuous variables are presented as mean± standard deviation, while binary variables are presented as number of observations and percentages. Groups were compared using Student's t-test for continuous variables and the chi-square test for categorical variables. Correlates of peak exercise systolic BP were identified by Pearson's correlation coefficient. Multiple linear regression analysis was used to identify independent covariates of peak systolic exercise BP, reporting standardized correlation coefficient (β) and p-value for individual variables. Logistic regression analysis was used to test the association between known hypertension and high systolic exercise BP and reported as odds ratio (OR) and 95% confidence interval (CI). A p < 0.05 was considered statistical significant both in uni- and multivariate analysis.

Results

The total study population of 77 subjects included 46 women (60%) was on average 48 ± 10 years old with mean BMI 32.6 ± 4.8 kg/m² and clinic systolic and diastolic BP 132 ± 17 mmHg and 82 ± 8 mmHg, respectively. Forty-two percent of the subjects had known hypertension, of which 59% were using antihypertensive drug treatments, on average 1.4 drugs. In the total study population, high systolic exercise BP was found twice as often in subjects with known hypertension (47% vs 22%, p < 0.05).

Compared with subjects with normal systolic exercise BP, those with high systolic exercise BP (peak exercise systolic BP ≥ 200 mmHg) had significantly higher clinic systolic BP (142 ± 20 mmHg vs 127 ± 13 mmHg, p < 0.01) and included more subjects with known hypertension (60% vs 33%, p < 0.05), while no difference in age, BMI or diastolic BP was found (Table II). The group with high systolic exercise BP also had significantly higher systolic 24-h ABP (126 ± 11 mmHg vs 120 ± 12 mmHg, p < 0.05) as well as higher daytime systolic ABP (130 ± 11 mmHg vs 123 ± 12 mmHg, p < 0.05), while night-time systolic ABP did not differ (Table III). Based on combined clinic BP and ABP, masked hypertension was present in six (13%) subjects without previous known hypertension. Of these, 50% had high systolic exercise BP.

Table II. Clinical characteristics and cardiovascular risk factors in groups of patients with and without high systolic exercise BP (≥ 200 mmHg).

	Normal systolic exercise BP (n = 52)	High systolic exercise BP (n = 25)
Women (%)	72	28
Age (years)	47.3 ± 9.8	50.6 ± 9.3
Weight (kg)	95.1 ± 16.5	100.7 ± 20.0
BMI (kg/m^2)	32.5 ± 4.8	33.0 ± 4.8
Hypertension (%)	33	60*
Diabetes mellitus (%)	10	4
Current smoking (%)	37	44
Family history of premature coronary heart disease (%)	37	36
Obstructive lung disease (%)	21	28
Fasting s-glucose (mmol/l)	5.4 ± 0.9	5.8 ± 1.0
Fasting s-insulin (pmol/l)	108 ± 141	108 ± 100
Triglycerides (mmol/l)	1.5 ± 1.1	1.6 ± 0.9
Total cholesterol (mmol/l)	5.4 ± 1.0	5.5 ± 1.0
HDL-cholesterol (mmol/l)	1.2 ± 0.3	1.4 ± 0.8
LDL-cholesterol (mmol/l)	3.5 ± 0.9	3.5 ± 0.8
Serum creatinine (μmol/l)	70 ± 12	68 ± 12
Hemoglobin (g/dl)	14.4 ± 1.3	14.4 ± 1.4
Clinic systolic BP (mmHg)	126.7 ± 13.2	142.0 ± 20.4**
Clinic diastolic BP (mmHg)	80.4 ± 8.7	83.8 ± 6.8
Clinic pulse pressure (mmHg)	46 ± 7	58 ± 17**
Clinic mean BP (mmHg)	96 ± 10	103 ± 10**
Clinic heart rate (beats/min)	67.5 ± 10.0	70.0 ± 7.0

*p < 0.05; **p < 0.01.

Table III. Ambulatory blood pressure (ABP) in groups with normal and high systolic exercise blood pressure (BP; ≥ 200 mmHg).

	Normal systolic exercise BP (n = 52)	High systolic exercise BP (n = 25)
24-h systolic ABP (mmHg)	119.5 ± 11.5	125.5 ± 10.7*
24-h diastolic ABP (mmHg)	73.0 ± 7.3	75.7 ± 8.1
24-h heart rate (beats/min)	73.0 ± 9.4	75.0 ± 7.9
Daytime systolic ABP (mmHg)	123.2 ± 12.0	129.8 ± 11.4*
Daytime diastolic ABP (mmHg)	75.8 ± 7.7	78.6 ± 8.1
Daytime heart rate (beats/min)	75.6 ± 10.1	77.7 ± 8.2
Night-time systolic ABP (mmHg)	106.7 ± 12.2	105.8 ± 24.9
Night-time diastolic ABP (mmHg)	63.9 ± 6.6	62.1 ± 15.9
Night-time heart rate (beats/min)	63.4 ± 8.7	62.6 ± 14.8

*p < 0.05.

Systolic and diastolic BP were significantly higher in the group with high systolic exercise BP both before exercise and after completion of exercise testing, while heart rate did not differ (Table IV). The group with high systolic exercise BP also had significantly lower peak oxygen uptake (28.1 ± 6.0 vs 31.7 ± 8.6 ml/kg/min, p < 0.05) compared with the group with normal systolic exercise BP despite comparable mean exercise time respiratory exchange ratio (RER) (Table IV).

Table IV. Blood pressure (BP) response and physical fitness during treadmill exercise.

	Normal systolic exercise BP (n = 52)	High systolic exercise BP (n = 25)
Systolic BP (mmHg)
Before exercise	133.3 ± 15.8	149.7 ± 18.1**
Peak exercise	171.0 ± 15.6	223.3 ± 17.6**
1 min post exercise	181.2 ± 25.2	207.4 ± 30.0**
5 min post exercise	144.8 ± 17.8	162.7 ± 24.5**
Diastolic BP (mmHg)
Before exercise	84.2 ± 10.0	89.8 ± 10.5*
Peak exercise^a	N/A	N/A
1 min post exercise	87.6 ± 11.4	97.4 ± 15.3*
5 min post exercise	85.1 ± 7.8	89.5 ± 9.2*
Heart rate (beats/min)
Before exercise	85.6 ± 12.7	86.6 ± 11.7
Peak exercise	176.1 ± 22.4	178.2 ± 20.6
Exercise duration (min)	15.3 ± 2.3	14.9 ± 1.8
Peak oxygen consumption (ml/kg/min)	31.7 ± 8.6	28.1 ± 6.0*
Respiratory exchange ratio	1.24 ± 0.17	1.26 ± 0.10
Metabolic equivalents	9.0 ± 2.6	8.3 ± 2.4

^aOnly systolic BP was measured during exercise. *p < 0.05; **p < 0.01.

In univariate analysis, higher exercise systolic BP correlated with higher clinic systolic BP (r = 0.41) (Figure 1), clinic diastolic BP (r = 0.31), 24-h systolic ABP (r = 0.32) and daytime systolic ABP (r = 0.32, all p < 0.01) and with lower peak oxygen uptake (r = − 0.25, p < 0.05). Higher age had a borderline significant association with higher systolic exercise BP (r = 0.22, p = 0.059). In univariate logistic regression known hypertension predicted a threefold increased risk of high systolic exercise BP (OR = 3.09, 95% CI 1.15–8.29; p = 0.025).

Figure 1. Correlation between clinic systolic blood pressure (BP) and systolic exercise BP in normotensive and hypertensive subjects with increased body mass index.

Based upon the univariate associations, variables were chosen for the multivariate model. Multiple linear regression analysis demonstrated that known hypertension (β = 0.33, p < 0.01), higher 24-h systolic ABP (β = 0.22, p < 0.05) and high-density lipoprotein (HDL)-cholesterol level (β = 0.23, p < 0.05) contributed significantly to prediction of higher exercise systolic BP, explaining 32% of the variability in peak exercise systolic BP (multiple R² = 0.32, p < 0.001), independent of gender and peak oxygen uptake. Addition of other significant factors from univariate analysis did not contribute to improving the yield of the model significantly.

Discussion

The present study demonstrates that among overweight and obese subjects, known hypertension, higher systolic ABP and HDL-cholesterol level were the strongest determinants of peak systolic BP during maximal exercise testing on treadmill. In our study, the CAEP protocol was chosen over the Bruce protocol due to the anticipated fitness level of the study subjects, as the CAEP is less demanding and therefore allows most subjects to complete several stages of exercise. Differences between the groups with normal and high systolic exercise BP were similar to those shown by Ilia et al. (16) in healthy normotensive subjects. In particular, subjects with high systolic exercise BP were older, included more men and had higher clinic BP, while no difference was found in exercise test duration. In contrast to Ilia et al., however, 42% of our overweight population had known hypertension. The lack of association between measures of obesity and exercise BP was unexpected, but probably reflects the limited BMI range included in the study.

In our study, high systolic exercise BP was found twice as often in subjects with known hypertension. This observation may be important, as Kjeldsen et al. (9) reported that men with both systolic BP ≥ 140 mmHg at rest and systolic exercise BP ≥ 200 mmHg had a 21-year cardiovascular death rate of 21.1% compared with 9.7% in those who remained below the abovementioned cut-offs. It is worthy of note that different exercise protocols were used in the studies, Kjeldsen et al. measuring systolic exercise BP after 6 min bicycling starting directly at 100 W workload, compared with the gradually increasing treadmill exercise used in the present study. Compared with clinic systolic BP, systolic ABP and exercise BP have also been found to be closed associated with increased left ventricular mass, a strong predictor of cardiovascular morbid events or death (6), probably reflecting that ABP more accurately measures left ventricular pressure overload.

The finding that known hypertension is associated with higher systolic exercise BP in our obese population is in line with a previous publication by Tsioufis et al. (17), demonstrating that newly diagnosed hypertensive patients with the metabolic syndrome had a 2.3-fold increased prevalence of exaggerated BP response during exercise testing. Pathophysiological mechanisms explaining these findings include increased arterial stiffness (18), increased sympathetic tonus (19) and abnormal vasoconstriction of muscular arteries during exercise, particularly in sedate hypertensive patients (20).

Ambulatory BP recording makes it possible to reveal masked hypertension, hence identifying hypertension in those whom diagnosis is missed by clinic BP measurement (21). Sharman et al. (22) found masked hypertension to be highly prevalent (58%) in apparently healthy subjects with normal clinic BP who had a hypertensive response to exercise. In their study, masked hypertension was also the strongest independent determinant of left ventricular mass index, a well-known predictor of cardiovascular morbidity and mortality (6,23). The present study did not include echocardiography. However, masked hypertension was found in 13% of subjects identified as normotensive by clinic BP, among whom 50% had high systolic exercise BP, comparably with what was found among subjects with known hypertension. The finding that ABP, but not clinic BP, predicted systolic exercise BP in our multivariate analysis may reflect that ABP is measured also while the patient is subject to activity and similar mechanisms that also produce an exaggerated BP response during exercise testing (6).

In the present study, HDL-cholesterol level was numerically, but not statistically higher in subjects with high systolic exercise BP. Hiratsuka et al. (24) recently reported that subjects with the metabolic syndrome and high levels of HDL-cholesterol (≥ 2.3 mmol/l) had increased insulin resistance, which was associated with increased BP response to exercise in hypertensive patients in a study by Park et al. (25). In another study among young military adults, high-normal BP was associated with low levels of HDL-cholesterol (< 1.0 mmol/l) (26), two traits commonly found in overweight population. Larger systolic BP response to exercise has previously been found to be frequent among individuals with high-normal BP (7).

An interesting finding in our study was that the group with high systolic exercise BP had similar night-time systolic BP, diastolic ABP and heart rate compared with the group with normal systolic exercise BP. Theoretically, this may reflect a higher sympathetic nervous system activity in the group with high systolic exercise BP, which is the main contributor to the diurnal variation in BP, or higher insulin resistance, another major stimulus for sympathetic nervous system activity (27,28). However, neither sympathetic nervous activity nor insulin resistance was measure in the present study.

Limitations

The main study limitation is the limited number of participants, but also lack of echocardiography and measurement of sympathetic nervous activity and insulin resistance, which might have proved helpful for additional explanation of the study findings. Our study was performed in overweight and obese subjects, and results should be generalized to other populations with caution.

Conclusion

In conclusion, our results demonstrate that history of hypertension, higher systolic ABP and HDL- cholesterol level were independent predictors of high systolic exercise BP in overweight and obese subjects. In overweight subjects with high systolic exercise BP, ABP monitoring should be performed to unmask hypertension.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.