Association of Serum Carotenoid Levels With N-Terminal Pro-Brain-Type Natriuretic Peptide: A Cross-Sectional Study in Japan

Background Several epidemiologic studies have reported an inverse association between serum levels of carotenoids and cardiovascular disease risk. However, no studies have reported an association between serum carotenoids and N-terminal pro-brain-type natriuretic peptide (NT-proBNP) in the general population. Methods In this cross-sectional study, we investigated whether serum carotenoids were associated with serum NT-proBNP in 1056 Japanese subjects (390 men, 666 women) who attended a health examination. Serum levels of carotenoids were separately determined by high-performance liquid chromatography. Serum NT-proBNP level was measured by electrochemiluminescence immunoassay. Results Serum NT-proBNP was elevated (≥55 pg/ml) in 31.8% of men and 48.2% of women. Multivariate logistic regression analyses adjusted for confounding factors showed a significant association between the highest quartile of serum α-carotene and elevated NT-proBNP in men (odds ratio [OR] = 0.40, 95% CI = 0.19–0.82, P for trend = 0.005) and women (OR = 0.62, 95% CI = 0.39–0.99, P for trend = 0.047). In women, moreover, elevated serum NT-proBNP was significantly associated with serum canthaxanthin (OR = 0.57, 95% CI = 0.36–0.90 for highest quartile, P for trend = 0.026) and β-cryptoxanthin (OR = 0.53, 95% CI = 0.32–0.85 for highest quartile, P for trend = 0.026), after adjusting for potential confounders. Conclusions Higher levels of serum carotenoids were associated with lower risk of elevated serum NT-proBNP levels after adjusting for possible confounders, which suggests that a diet rich in carotenoids could help prevent cardiac overload in the Japanese population.


INTRODUCTION
Brain-type natriuretic peptide (BNP) is secreted predominantly from the ventricular myocardium in response to increased ventricular wall stretch. 1 In the process of BNP secretion, pro-BNP in cardiomyocytes is cleaved into the active hormone BNP and the inactive N-terminal pro-BNP (NT-proBNP).][7] Fruits and vegetables contain more than 40 carotenoids routinely absorbed and metabolized by humans. 8Circulating carotenoids are regarded as useful biomarkers of total dietary intake of vegetables and fruits. 91][12] Carotenoids have various biologic functions, including antioxidant antiinflammatory effects. 13,14Carotenoids may reduce CVD risk, in part because of their antioxidant and anti-inflammatory properties. 15xidative stress is associated with HF pathogenesis.Experimental and clinical studies have shown that generation of reactive oxygen species (ROS) is increased in HF. 16,17 Several small case-control studies have reported that circulating levels of antioxidants, such as β-carotene and vitamin E, were lower in HF patients than in control subjects and that the reductions correlated with the severity of cardiac overload. 18,19Although some studies have reported associations between carotenoids and cardiovascular complications, no epidemiologic studies have evaluated the associations between serum levels of carotenoids and NT-proBNP, a biomarker of cardiac load.We investigated whether circulating carotenoids were independently associated with NT-proBNP in the Japanese general population.

METHODS
Health examinations of inhabitants aged 39 years or older have been performed in Yakumo Town, Hokkaido Prefecture, Japan, since 1982. 20The initial population in the present study comprised 1178 adults (428 men, 750 women) who attended health examinations in August 2003 or August 2004.Among these 1178 adults, we excluded 28 who had serum samples that were inadequate for measurement of NT-proBNP or carotenoid levels and 94 with heart disease or kidney disease.Ultimately, data from 1056 subjects (390 men, 666 women) were analyzed.All study protocols were approved by the ethics committee of Fujita Health University (approval number 11-101).
Fasting serum samples were collected during health examinations, and sera were separated from blood cells by centrifugation within 1 hour.Serum samples were stored at −80°C until analyzed for carotenoids and NT-proBNP.Serum carotenoids and NT-proBNP levels were measured by the end of the year in which the sera were collected (2003 or  2004).NT-proBNP and carotenoids remain stable if the serum is frozen at −80°C until analyzed. 21,22Serum levels of carotenoids were separately determined by high-performance liquid chromatography. 23The coefficients of variation for the repeatability and reproducibility of the assay of carotenoids were 4.5% to 9.2% and 9.2% to 20.0%, respectively. 23Serum NT-proBNP level was measured by electrochemiluminescence immunoassay (Roche Diagnostics, Tokyo, Japan).The withinrun coefficient of variation ranged from 1.3% to 2.4% and the between-run coefficient of variation ranged from 2.9% to 6.1% for the NT-proBNP assay. 24Other biochemical analyses of blood were performed using autoanalyzers in the laboratory at Yakumo General Hospital.
Trained nurses administered a questionnaire on health and daily lifestyle habits, smoking (current smoker, ex-smoker, or non-smoker), alcohol consumption (regular drinker, exdrinker, or non-drinker), and history of major illness.Anthropometric indices (height, weight, and waist and hip circumferences) and blood pressure were measured during the health examination.Body mass index (BMI) was calculated as weight divided by the height squared (kg/m 2 ).Estimated glomerular filtration rate (eGFR) was used to assess kidney function, according to the following equation: eGFR (ml/min/1.73m 2 ) = 194 × creatinine −1.094 × age −0.287 (× 0.739 if female). 25ll statistical analyses were conducted using JMP version 9.0 software (SAS Institute, Cary, NC, USA).Because serum levels of carotenoids and NT-proBNP are distributed logarithmically, we used logarithms of these variables in the analyses.Categorical variables were compared by the chisquare test; continuous variables were analyzed using the ttest.Relationships between levels of serum carotenoids and NT-proBNP were evaluated by partial correlation analyses.Logistic regression analysis was used to estimate odds ratios (ORs) with 95% CIs for elevated NT-proBNP, adjusted for potential confounders.Elevated NT-proBNP was defined as a level of 55 pg/ml or higher, based on the findings of a previous report. 268][29][30][31] A P value of less than 0.05 was considered statistically significant.

RESULTS
Mean age was 61.9 years (range: 39-87 years) in men and 59.4 years (range: 39-86 years) in women.Serum NT-proBNP was significantly higher in women than in men.Serum levels of carotenoids, including zeaxanthin/lutein, canthaxanthin, β-cryptoxanthin, lycopene, α-carotene, and β-carotene, were all significantly higher in women than in men.Table 1 shows a comparison of the characteristics of subjects with elevated (≥55 pg/ml) and low serum NT-proBNP.The proportion of subjects with elevated serum NT-proBNP was 31.8% in men and 48.2% in women.Mean age and SBP were significantly higher in subjects with high NT-proBNP than in those with low NT-proBNP.Serum levels of β-cryptoxanthin and lycopene and eGFR were significantly lower in subjects with high NT-proBNP than in those with low NT-proBNP.
The multivariable-adjusted partial correlation coefficients between serum NT-proBNP level and serum levels of carotenoids are shown in Table 2. Serum NT-proBNP was significantly inversely associated with serum levels of canthaxanthin, lycopene, and α-carotene among women.No significant associations between serum NT-proBNP and serum levels of carotenoids were observed among men after adjustment for potential confounders.
We calculated the OR and 95% CI for elevated serum NT-proBNP, adjusted for confounding factors, by quartile of serum carotenoids (Table 3).In women, the OR for elevated serum NT-proBNP was significantly lower in the highest  quartiles of serum canthaxanthin, β-cryptoxanthin, and αcarotene levels than in the lowest quartile.Among men, the OR was significantly lower in the highest and third quartiles of serum α-carotene levels as compared with the lowest quartile.Moreover, ORs for elevated serum NT-proBNP tended to decrease as serum canthaxanthin and β-cryptoxanthin increased in men (P for trend = 0.054 and 0.059, respectively).

DISCUSSION
Serum levels of several carotenoids were inversely associated with serum NT-proBNP level in Japanese women, even after adjusting for possible confounding factors.Weak inverse associations were seen in Japanese men.These results suggest that high intake of vegetables and fruits ameliorates cardiac overload.
We observed no association between serum α-carotene and serum NT-proBNP in men (Table 2).However, the OR for elevated serum NT-proBNP decreased across increasing quartiles of serum α-carotene (Table 3).This discrepancy may be due to the difference in statistical methods.Partial correlation represents the strength of the relationship between 2 continuous variables after controlling for other variables.OR represents the strength of an association between 2 binary valuables.Although serum α-carotene was not linearly related with serum NT-proBNP, higher serum α-carotene was associated with a lower risk of elevated NT-proBNP (≥55 pg/ml) in men.
2,33 BNP is predominantly produced in cardiac tissue, 1 and the left ventricle is the primary source of circulating BNP in the normal state and under conditions of left ventricular dysfunction. 32,33The BNP gene is activated in cardiomyocytes in response to increased stress of the myocardial wall.Thus, precursor proBNP is produced intracellularly and then cleaved by endoprotease upon secretion, which results in the formation of biologically inert NT-proBNP and biologically active BNP. 34BNP assists in regulating blood pressure, blood volume, and sodium balance. 35][7] Increased oxidative stress contributes to HF pathogenesis.Clinical studies have reported that chronic HF was associated with increased plasma oxidants, such as malondialdehyde and lipid peroxides, and decreased plasma levels of antioxidant vitamins, including vitamins C and E and β-carotene. 18,19,36 large prospective study found that plasma vitamin C was inversely associated with HF risk in a healthy European population. 37OS are highly reactive molecules that can oxidize DNA, proteins, and lipids.They are produced by several mechanisms such as mitochondrial electron transport, NAD(P)H oxidase, and xanthine oxidase within cardiac myocytes.ROS cause contractile failure and structural damage in the myocardium. 38xidative stress is caused by an imbalance between ROS generation and the antioxidant defense system.Increased oxidative stress also indicates enhanced ROS production.Therefore, decreased circulating levels of antioxidants such as vitamin C and carotenoids may increase ROS.
Carotenoids are potent quenchers of singlet molecular oxygen 14 and, by quenching free radicals, have an important role in decreasing activation of proinflammatory pathways. 13ietary intake of carotenoids was found to be associated with decreased CVD risk. 15,39,40Plasma concentrations of carotenoids are believed to be useful biomarkers of total dietary intake of fruits and vegetables. 9Several prospective epidemiologic studies have shown that high serum levels of carotenoids are associated with low CHD risk. 10,12,41A cohort study by Morris et al 12 found that high serum levels of total carotenoids were associated with lower risk of incident CHD in men.The Basel Prospective Study 41 found that the risk of ischemic heart disease and stroke was significantly increased in subjects with initially low plasma levels of αand βcarotene.In addition, a Japanese prospective study reported that high serum levels of αand β-carotene were significantly associated with low CVD risk. 10Antioxidant and antiinflammatory properties may explain the possible role of carotenoids in HF and CVD prevention.Several limitations of this study warrant mention.First, although some epidemiologic studies have reported an inverse association of circulating levels of carotenoids with CVD risk, 10,12,41 large randomized trials have found β-carotene supplementation to be ineffective in preventing CVD. 42,43ecause high doses of carotenoids could have a pro-oxidant effect, it is possible that the effects of carotenoids in protecting against CVD disappear at the high doses used in supplementation studies.Supplementation with β-carotene had little effect on our results, because only 3 men and 9 women in this study used β-carotene supplements.][46] We conducted sex-specific analyses because women had NT-proBNP levels that were approximately 1.4 times those of men in this study, as was the case in previous studies. 44,45The age-related increase in NT-proBNP may reflect the higher prevalence of subclinical cardiac disease and renal dysfunction in older subjects. 47Because NT-proBNP is mainly cleared by the kidneys, circulating levels are often elevated in patients with renal dysfunction.The possibility of residual confounding cannot be completely excluded, although other confounding variables were appropriately adjusted for in our analyses.Finally, this study was unable to examine issues of causality, due to the cross-sectional design.A prospective study is thus required to confirm our results and clarify causal relationships.
In conclusion, levels of serum carotenoids, which are markers of fruit and vegetable intake, were inversely associated with risk of elevated serum NT-proBNP in the general Japanese population, although serum levels of all carotenoids were not linearly related to serum NT-proBNP.These findings suggest that a diet rich in vegetables and fruits prevents cardiac overload and thus reduces CVD risk.

ONLINE ONLY MATERIAL
Abstract in Japanese.

Table 2 .
Relationships of serum NT-proBNP with serum retinol and serum carotenoids by sex

Table 3 .
Adjusted a odds ratios for elevated serum NT-proBNP according to serum carotenoid levels Adjusted for age, body mass index, smoking habit, drinking habit, estimated glomerular filtration rate, total cholesterol, γ-glutamyl transpeptidase, hemoglobin A1c, and systolic blood pressure.Data are expressed as odds ratio (95% CI).Elevated NT-proBNP was defined as a level ≥55 pg/ml. a