Cross-sectional studies have reported inverse associations of B-type natriuretic peptide (BNP) with the body mass index (BMI). We evaluated whether changes in the BMI are associated with changes in BNP. A nested prospective cohort study of a lifestyle intervention (low-fat, whole-foods diet, exercise, stress management, and social support) was conducted. BNP, BMI, and other biomarkers were measured at baseline and 3 months. A total of 131 subjects, 56 with coronary heart disease (CHD) and 75 at high risk, with ≥3 CHD risk factors and/or diabetes mellitus, were enrolled. At 3 months, the mean BMI had decreased (34.4 to 31.7 kg/m 2 , p <0.001), BNP had increased (median 18 to 28 pg/ml, p <0.001), and low-density lipoprotein, C-reactive protein, apolipoprotein B (all p <0.002), and angina frequency (p = 0.017) and severity (p = 0.052) had decreased. The subjects’ physical limitations had decreased and their physical functioning had improved (all p <0.001). The percentage of change in BNP was inversely associated with the percentage of change in insulin (r = −0.339, p = 0.005, n = 63 nondiabetics). It was also inversely associated with the percentage of change in BMI (r = −0.28, p = 0.002, n = 116), and this association remained significant (p = 0.029) in multiple regression analyses controlling for age, gender, CHD, diabetes mellitus, percentage of change in lifestyle index, and β-blocker use. The metabolic changes related to adipose tissue lipolysis could explain these findings. In conclusion, BNP increased in subjects experiencing weight loss while following a lifestyle intervention, and angina pectoris, physical limitations, and other CHD risk factors decreased. Therefore, in this context, increasing BNP might not indicate worsening disease or a worsening prognosis. Thus, the proposed use of BNP in monitoring disease progression should take into account changes in the BMI during the same period.
Circulating B-type natriuretic peptide (BNP) levels aid in the diagnosis of acute congestive heart failure (HF) and might be useful in establishing the prognosis of patients with HF, with increased levels predictive of increased mortality. Periodic BNP measurement has been proposed for the purpose of monitoring the response to therapy for HF and for its prognostic value. However, BNP levels are influenced by other factors such as age, gender, diabetes, and impaired renal function. In addition, an inverse association of BNP level with the body mass index (BMI) has been reported in cross-sectional studies. Also, BMI greatly influences the sensitivity of BNP in the diagnosis of acute HF. A study of 22 patients, who had undergone bariatric surgery for obesity, found that during the subsequent 6 months, the BMI had decreased significantly and the BNP and N-terminal pro BNP levels increased significantly. The present report describes the changes in BNP in relation to changes in BMI in a cohort of 125 patients who underwent lifestyle changes during a 3-month period and showed improvement in other cardiac risk factors.
Methods
A prospective cohort study nested within a larger cohort participating in the health insurance administered Multisite Cardiac Lifestyle Intervention Program was conducted. The participating hospital sites included the Charleston Area Medical Center and West Virginia University Hospitals (West Virginia) and the Hamot Medical Center, Jameson Health System, and Jefferson Regional Medical Center (Pennsylvania). All the sites provided institutional review board approval, and each participant provided written informed consent before enrollment. The present study was registered at the Clinicaltrials.gov Website ( NCT00820313 ).
The details on the eligibility criteria for the Multisite Cardiac Lifestyle Intervention Program and the lifestyle intervention have been previously published. In brief, the inclusion criteria were coronary heart disease (CHD) or type 1 or type 2 diabetes or a high-risk status for CHD. The diagnosis of CHD was determined by one of the following criteria: (1) noninvasive testing demonstrating myocardial ischemia, (2) cardiac catheterization, (3) eligibility for coronary artery bypass grafting/percutaneous transluminal coronary angioplasty and seeking a clinical alternative, or (4) a history of coronary artery bypass grafting/percutaneous transluminal coronary angioplasty/stent implantation or myocardial infarction. The criteria for the high-risk category were (1) a family history of premature CHD (first-degree relative [men <55 years old and women <65 years old] with myocardial infarction or sudden cardiac death); or (2) men aged >45 years (women aged >55 years) with ≥2 of the following risk factors: current cigarette smoking (within the past 5 years), hypertension (blood pressure >140/90 mm Hg or taking antihypertensive medication), a low high-density lipoprotein cholesterol level (<35 mg/dl) or taking lipid-lowering medications, elevated lipoprotein (a) >30 mg/dl, total cholesterol >240 or taking lipid-lowering medications, low-density lipoprotein cholesterol >160 mg/dl or taking lipid-lowering medications, high-sensitivity C-reactive protein level of 3 to 10 mg/L, BMI >30 kg/m 2 , and/or insulin-resistant state (metabolic syndrome X). The primary exclusion criteria included (1) ischemic left main CHD, with obstruction >50%; (2) >70% proximal left anterior descending artery disease and proximal left circumflex artery disease and an ejection fraction of <50%; (3) unstable angina pectoris; (4) a history of exercise-induced ventricular tachycardia or third-degree heart block without evidence of current stability; (5) coronary artery bypass grafting or myocardial infarction within the previous 4 weeks; (6) HF with functional limitations; (7) current tobacco use; (8) uncontrolled malignant ventricular arrhythmia; and (9) impaired cognitive function (ie, dementia or delirium).
The guidelines for the lifestyle intervention included approximately 10% of daily calories from fat, 15% from protein, and 75% from complex carbohydrates. The exercise portion included a minimum of 3 hours per week of aerobic exercise, with a minimum of 30 minutes per session exercising within the prescribed target heart rates and/or perceived exertion levels, strength training activities a minimum of 2 times per week. Also, the participants were to practice stress management techniques for ≥1 hour per day. Finally, the participants were required to attend weekly group support sessions led by a licensed mental health professional twice each week.
The demographic information and medical history were obtained at baseline by interview and a review of the medical records. Clinical measurements, blood test results, and questionnaires were collected at baseline and at 3 months. A fasting blood sample was drawn for laboratory analyses, which included total cholesterol, high-density lipoprotein cholesterol, triglycerides, low-density lipoprotein cholesterol, high-sensitivity C-reactive protein, fibrinogen, lipoprotein (a), homocysteine, oxidized low-density lipoprotein, insulin, BNP, and nuclear magnetic resonance Lipoprofile assays for very-low-density lipoprotein, low-density lipoprotein, and high-density lipoprotein particle concentrations and particle size. In addition, the fasting blood glucose and hemoglobin A1c levels were tested for all patients with diabetes and for a subset of patients without diabetes. Self-administered questionnaires were completed by the participants to assess their exercise and stress management duration and frequency per week, psychosocial factors, and cognitive function. The angina frequency and angina severity were assessed. The participants also completed 3-day food diaries. These were entered into the software program Food Processor, version 10x (ESHA Research, Salem, Oregon) by registered dieticians for nutrient analysis.
Data analysis was conducted using the Statistical Package for Social Sciences, version 14.0 (SPSS, Chicago, Illinois). Continuous data are presented as the mean ± SD for normally distributed variables and as the median and interquartile range or range for non-normally distributed variables. For significance testing, t tests and the Mann-Whitney rank sum test were used. Chi-square tests were used to evaluate differences in proportions between groups. A 2-way independent analysis of variance was conducted to evaluate differences in baseline BNP levels by gender and CHD status and to evaluate the interaction of gender and CHD status. Changes from baseline were tested for significance using paired t tests for normally distributed variables and the Wilcoxon signed rank test for non-normal variables. A diet adherence score and a lifestyle adherence score were calculated using the same formula as previously described. Higher scores indicated better adherence to the recommendations. The association between the continuous variables was evaluated using linear regression analysis. Pearson’s r and associated 2-sided p values were computed. The BNP and insulin values had a non-normal distribution, and a log transformation for the baseline and 3-month values was done. Multiple linear regression analyses were conducted with the change scores for log transformed BNP values (3-month value [time 2] minus the baseline value [time 1]: log 10 [time 2/time 1]) as the dependant variable and age, gender, CHD, diabetes, percentage of change in BMI, percentage of change in lifestyle index, and β-blocker use as independent variables. The change in insulin was considered likely to be on the pathway of the association between the change in BNP and the change in BMI; therefore, a separate regression model was created with the change score log BNP as the dependant variable and the change score log insulin, β-blocker use, and insulin*β blocker as independent variables to evaluate the effect modification of this association by β-blocker use. The percentage of change in BMI was recoded as an ordinal variable with 4 categories. A 1-way analysis of variance test was done to test for differences in the percentage of change in BNP between these categories, using Bonferroni correction for multiple comparisons.
Results
A total of 131 participants (59.2% women and 43.1% with diabetes), 56 with pre-existing CHD (37.5% women and 27.3% with diabetes) and 75 at high risk of CHD with ≥3 CHD risk factors and/or diabetes (76% women and 54.7% with diabetes) were enrolled from January 2007 to March 2008. Of the 131 participants, 6 participants had withdrawn from the study by the 3-month follow-up point. The baseline patient characteristics are summarized in Table 1 , and a subset of variables measured at baseline and at 3 months is summarized in Table 2 .
| Characteristic | Diagnosed With CHD (n = 54) | High Risk of CHD (n = 71) | ||
|---|---|---|---|---|
| Men (n = 35) ⁎ | Women (n = 19) ⁎ | Men (n = 16) ⁎ | Women (n = 55) ⁎ | |
| Age (years) | 58.2 ± 7.6 | 59.9 ± 7.0 | 57.3 ± 7.8 | 56.1 ± 9.8 |
| White | 32/35 (91%) | 19/19 (100%) | 15/16 (93%) | 51/55 (92%) |
| Previous cigarette smoker | 15/35 (43%) | 5/19 (26%) | 8/16 (50%) | 19/55 (34%) |
| Diabetes mellitus | 8/35 (22%) | 7/19 (36%) | 10/16 (62%) | 29/55 (52%) |
| Medications | ||||
| Nitrate | 3/35 (8%) | 1/19 (5%) | 0/16 (0%) | 0/55 (0%) |
| β Blocker | 25/35 (71%) | 12/19 (63%) | 3/16 (18%) | 12/55 (22%) |
| Angiotensin-converting enzyme inhibitor | 16/35 (46%) | 5/19 (26%) | 4/16 (25%) | 9/55 (16%) |
| Lipid-lowering medication | 32/35 (91%) | 16/19 (84%) | 9/16 (56%) | 20/55 (36%) |
| Antiplatelet | 34/35 (97%) | 16/19 (84%) | 6/16 (37%) | 15/55 (27%) |
| Body mass index (kg/m 2 ) | 32.14 ± 6.23 | 32.4 ± 6.6 | 36.5 ± 6.8 | 35.4 ± 7.0 |
| Total cholesterol (mg/dl) | 157 (130–189) | 171 (154–202) | 167 (153–209) | 191 (170–228) |
| Low-density lipoprotein cholesterol (mg/dl) | 90 (66–110) | 89 (73–106) | 91 (76–126) | 112 (92–139) |
| High-density lipoprotein cholesterol (mg/dl) | 39 (33–45) | 44 (38–61) | 42 (39–47) | 45 (37–50) |
| B-type natriuretic peptide (pg/ml) | 30 (13–50) | 27 (16–94) | 15 (8–20) | 16 (9–27) |
| Insulin (μU/L) | 13 (9–24) | 13 (9–26) | 18 (12–33) | 15 (9–24) |
⁎ Because of missing data, n for individual variables ranged from 34 to 35 for men with CHD, 17 to 19 for women with CHD, 15 to 16 for high-risk men, and 53 to 55 for high-risk women.
| Variable | Baseline | 3 Months | p Value |
|---|---|---|---|
| Fat intake (% total calories) | 29.1 (21.0–35.8) | 10.5 (9.0–12.1) | <0.001 ⁎ |
| Dietary cholesterol (mg) | 152 (91–250) | 7 (3–11) | <0.001 ⁎ |
| Dietary fiber (g) | 22 (14–30) | 34 (28–41) | <0.001 ⁎ |
| Exercise (minutes/week) | 0 (0-115) | 212 (180–275) | <0.001 ⁎ |
| Stress management † (minutes/week) | 0 (0-230) | 420 (0-630) | <0.001 ⁎ |
| Group session attendance (% classes attended) | NA | 100 (92–100) | NA |
| Body mass index (kg/m 2 ) | 34.2 ± 6.8 | 31.8 ± 6.2 | <0.001 ⁎ |
| Systolic blood pressure (mm Hg) | 127 ± 15 | 117 ± 11 | <0.001 ⁎ |
| Diastolic blood pressure (mm Hg) | 77 ± 8 | 72 ± 7 | <0.001 ⁎ |
| Functional capacity (METS) | 7.6 (5.6–10.0) | 10.1 (7.8–12.5) | <0.001 ⁎ |
| Angina frequency ‡ | 0.0 (0.0–5.0) | 0.0 (0.0–1.0) | 0.017 ⁎ |
| Angina severity ‡ | 0.0 (0.0–2.0) | 0.0 (0.0–3.0) | 0.052 |
| Total cholesterol § (mg/dl) | 180 (153–205) | 161 (143–198) | <0.001 ⁎ |
| Low-density lipoprotein § (mg/dl) | 103 (78–124) | 81 (64–112) | <0.001 ⁎ |
| High density lipoprotein § (mg/dl) | 43 (37–51) | 37 (33–44) | <0.001 ⁎ |
| Triglycerides § (mg/dl) | 129 (95–185) | 125 (80–176) | 0.070 |
| B-natriuretic peptide § (pg/ml) | 18.0 (11.0–35.0) | 28.0 (14.0–52.3) | <0.001 ⁎ |
| C-reactive protein § (mg/dl) | 2.0 (0.8–5.1) | 1.4 (0.6–3.6) | <0.001 ⁎ |
| Apolipoprotein B § (mg/dl) | 87 (69–109) | 84 (65–99) | 0.001 ⁎ |
| Insulin (μU/L) | 14.5 (9.7–25.2) | 12.0 (9.0–18.3) | <0.001 ⁎ |
| Role physical, ‡ SF-36 | 75 (25–100) | 100 (75–100) | <0.001 ⁎ |
| Physical functioning, ‡ SF-36 | 80 (55–90) | 90 (80–95) | <0.001 ⁎ |
| Physical component score, ‡ SF-36 | 47 (39–52) | 52 (46–56) | <0.001 ⁎ |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
