SEARCH

SEARCH BY CITATION

Keywords:

  • congestive heart failure;
  • fatty acids;
  • lipids

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Conflict of interest statement
  9. Acknowledgements
  10. References
  11. Supporting Information

Abstract.  Øie E, Ueland T, Dahl CP, Bohov P, Berge C, Yndestad A, Gullestad L, Aukrust P, Berge RK (Research Institute for Internal Medicine, Department of Cardiology, Oslo University Hospital, Rikshospitalet, Oslo, Center for Heart Failure Research, University of Oslo, Oslo, Section of Endocrinology, Oslo University Hospital, Rikshospitalet, Oslo, Faculty of Medicine, University of Oslo, Oslo, Section of Medical Biochemistry, Institute of Medicine, University of Bergen, Bergen, Department of Heart Disease, Haukeland University Hospital, University of Bergen, Bergen; and Section of Clinical Immunology and Infectious Diseases, Oslo University Hospital, Rikshospitalet, Oslo; Norway). Fatty acid composition in chronic heart failure: low circulating levels of eicosatetraenoic acid and high levels of vaccenic acid are associated with disease severity and mortality. J Intern Med 2011; 270: 263–272.

Objectives.  Free fatty acids (FFAs) are the major energy sources of the heart, and fatty acids (FAs) are active components of biological membranes. Data indicate that levels of FAs and their composition may influence myocardial function and inflammation. The aim of this study was to investigate whether total levels and composition of FAs and FFAs in plasma are altered in clinical heart failure (HF) and whether any alterations in these parameters are correlated with the severity of HF.

Subjects.  Plasma from 183 patients with stable HF was compared with plasma from 44 healthy control subjects.

Results.  Our main findings are as follows: (i) patients with HF had decreased levels of several lipid parameters and increased levels of FFAs in plasma, compared with controls, which were significantly correlated with clinical disease severity. (ii) Patients with HF also had a decreased proportion in the plasma of several n-3 polyunsaturated FAs, an increased proportion of several monounsaturated FAs, and a decreased proportion of some readily oxidized long-chain saturated FAs. (iii) These changes in FA composition were significantly associated with functional class, impaired cardiac function (i.e., decreased cardiac index and increased plasma N-terminal pro-B-type natriuretic peptide levels) and enhanced systemic inflammation (i.e., increased high-sensitivity C-reactive protein levels). (iv) Low levels of C20:4n-3 (eicosatetraenoic acid) and in particular high levels of C18:1n-7 (vaccenic acid) were significantly associated with total mortality in this HF population.

Conclusions.  Our data demonstrate that patients with HF are characterized by a certain FA phenotype and may support a link between disturbed FA composition and the progression of HF.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Conflict of interest statement
  9. Acknowledgements
  10. References
  11. Supporting Information

During the last decade, it has become evident that altered myocardial metabolism is a hallmark of heart failure (HF) [1–4]. In the foetal heart, glucose is the main energy source. However, free fatty acids (FFAs) are the main myocardial substrate (∼70%) postnatally (for review, see ref. [2]). Although FFAs provide the highest ATP yield, FFA metabolism requires more oxygen than glucose metabolism and is thus less energy efficient. Hence, high plasma FFA concentrations may be detrimental to the heart, increasing oxygen consumption for any given workload, and circulating FFA levels may be an important regulator of myocardial substrate metabolism in HF, potentially contributing to myocardial ischaemia and dysfunction.

Membrane phospholipids and their constituent fatty acids (FAs) are active components of biological membranes and play an important role in signal transduction and cellular metabolism. The composition of FAs may also be of importance for myocardial function [5, 6]. Hence, it has been reported that greater levels of saturated FAs may increase the risk of HF in apparently healthy individuals and, in the same population, arachidonic acid and n-3 polyunsaturated FAs (PUFAs) were found to decrease the risk of HF in women [6]. Also, in a large community-based cohort of Japanese men and women, there was an inverse association between fish and n-3 PUFA dietary intakes and cardiovascular death, especially from HF [7]. Moreover, n-3 PUFA supplementation was recently shown to decrease mortality and admission to hospital for cardiovascular reasons in patients with HF, further supporting a beneficial effect of these FAs on myocardial function [8]. However, whereas several studies have investigated the role of n-3 PUFAs in HF, few have examined the composition of other FAs in patients with HF in relation to disease severity.

To further elucidate the role of FAs and their composition in HF, we investigated whether the total spectrum of FA composition in plasma is altered in clinical HF and whether any alterations in these parameters are correlated with the severity of disease.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Conflict of interest statement
  9. Acknowledgements
  10. References
  11. Supporting Information

Study subjects

A total of 183 patients with stable symptomatic HF for >6 months admitted to our hospital for evaluation of the aetiology of HF and to assess whether they were candidates for cardiac resynchronization therapy or heart transplantation were consecutively included in the study (Table 1). In total, 176 patients were diagnosed with systolic HF (left ventricular ejection fraction <40%) and seven patients with HF with preserved ejection fraction, indicating diastolic HF. Patients with acute coronary syndromes during the last 6 months and those with significant concomitant disease, such as infection, malignancy, autoimmunity and gastrointestinal disorders, were not included. Forty-four healthy volunteers (based on self-reporting) were recruited as control subjects and were comparable to the patients with regard to age (patients with HF: 56 ± 13 years; controls: 57 ± 7 years, P = 0.70) and sex (patients with HF: 81% men; controls: 73% men, P = 0.30). The underlying cause of HF was classified as coronary artery disease (CAD, n = 71), dilated cardiomyopathy (DCM, n = 94) or other (n = 18) on the basis of disease history and coronary angiography. No information was available regarding intake of vitamins, fish oils, antioxidants or micronutrients in either patients or controls.

Table 1.   Clinical and haemodynamic characteristics of the heart failure (HF) population
 HF patients (n = 183)
  1. Data are presented as mean ± SD, or number or percentage of subjects. CAD, coronary artery disease; DCM, dilated cardiomyopathy; LVEF, left ventricular ejection fraction; PCWP, pulmonary capillary wedge pressure; ACE, angiotensin-converting enzyme; ARB, angiotensin II receptor blocker; NYHA, New York Heart Association.

Age (year)56 ± 13
Gender (male/female)148/35
Aetiology (CAD/DCM/other) (%)39/51/10
NYHA class (II/III/IV) (%)27/43/30
History (%)
 Diabetes mellitus14
 Hypertension16
 Previous myocardial infarction36
Biochemical values
 Creatinine (μmol L−1)101 ± 46
 Nt-proBNP (pmol L−1)437 ± 547
Haemodynamics
 LVEF (%)30 ± 12
 PCWP (mmHg)18 ± 8
 Cardiac index (L min−1 m−2)2.1 ± 0.6
Medication (%)
 ACE inhibitor66
 ARB20
 Beta-blocker77
 Diuretics68
 Aldosterone antagonist41
 Digitoxin27
 Statins42
 Warfarin45

Blood sampling protocol

Blood was collected from the cubital vein ∼2 h after a light, standard Norwegian breakfast. None of the individuals had been performing any exercise at the same day as the blood was collected. Blood was drawn into chilled glass tubes containing EDTA (1 mg mL−1 blood), placed on ice and centrifuged within 20 min at 4 °C (2000 g for 20 min). The plasma was immediately stored at −80 °C until use to avoid ex vivo oxidation or modification of FAs.

Measurements of plasma lipoproteins

Plasma lipids were measured on the Hitachi 917 system (Roche Diagnostics, Mannheim, Germany) using the following kits: total cholesterol (Bayer, Tarrytown, NY), free cholesterol (Wako Chemicals, Dalton, OH, USA), high-density lipoprotein (HDL) cholesterol (HDL-C Plus; Roche Diagnostics), low-density lipoprotein (LDL) cholesterol (LDL-C Plus; Roche Diagnostics), triacylglycerol (TG; Bayer) and phospholipids (PAP 150; BioMérieux, Lyon, France).

Measurements of plasma FFAs and total levels and composition of FAs

Plasma levels of FFAs were measured on the Hitachi 917 system using a commercially available FFA kit (NEFA C, Wako Chemicals, Neuss, Germany). Measurement of total levels and composition of FAs was performed after extracting lipids from plasma using a mixture of chloroform and methanol. The extracts were transesterified using BF3/methanol. Extracts of fatty acyl methyl esters were heated in 0.5 mol L−1 KOH in ethanol/water solution (9:1) to remove neutral sterols and nonsaponifiable material. Recovered FAs were re-esterified using BF3/methanol. The methyl esters were quantified by gas chromatography as previously described [9].

Measurements of N-terminal pro-B-type natriuretic peptide and high-sensitivity C-reactive protein

N-terminal pro-B-type natriuretic peptide (Nt-proBNP) in plasma was determined by an electrochemiluminescence immunoassay on a Modular platform (Roche Diagnostics, Basel, Switzerland). The plasma level of high-sensitivity C-reactive protein (hsCRP) was determined by a high-sensitivity particle-enhanced immunoturbidimetric assay [Tina-quant CRP (Latex) HS; Roche Diagnostic]. Cardiac index (CI) was measured during right-sided cardiac catheterization.

Ethics

The study was approved by the regional ethics committee, and the investigation conforms to the principles of the Declaration of Helsinki. All subjects gave written informed consent to participate in the study.

Statistical analysis

All the data are presented as mean ± SD. For comparisons of two groups, the Mann–Whitney U-test was used. anova with the Kruskal–Wallis test was used for comparisons of three or more groups. If the Kruskal–Wallis test revealed significant differences, subsequent analyses of individual group means were performed with the Mann–Whitney U-test. The correlation between plasma FFA and FA levels and New York Heart Association (NYHA) class, Nt-proBNP, CI and hsCRP was assessed by Spearman rank correlation test. Receiver operating characteristic (ROC) curves for associations between FAs and the end-point all-cause mortality or heart transplantation were generated and compared. Significant candidates were identified and tertiles calculated for all continuous data. The relationship between tertiles of plasma levels of FAs and long-term adverse events was visualized using Kaplan–Meier plots. Cox proportional hazards regression was used to calculate crude and adjusted risk estimates associated with tertiles in plasma levels of FAs for the end-point. In the multivariable analyses, we adjusted for statin use and traditional risk factors associated with HF mortality (age, hypertension, type 2 diabetes, creatinine, Nt-proBNP and hsCRP). Throughout, we report two-tailed P values, and values <0.05 were considered significant. However, particular attention should be directed towards smaller P values <0.01, because a considerable number of P values have been calculated.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Conflict of interest statement
  9. Acknowledgements
  10. References
  11. Supporting Information

Plasma lipid levels

The analysis of plasma levels of lipids in patients with HF and controls revealed several significant findings (Table S1). Patients with HF were characterized by significantly decreased levels of several lipid parameters such as total cholesterol, LDL cholesterol, HDL cholesterol, phospholipids and cholesterol esters. Not surprisingly, patients who were treated with statins had lower levels of most lipid parameters than non-statin users except for TG and HDL cholesterol. However, lipid levels were still significantly decreased compared with controls in patients with HF who were not receiving statin treatment. Patients with CAD had lower levels of several lipid parameters than those with DCM, possibly, at least in part, reflecting a higher proportion of statin users in this group (70% vs. 24% of DCM patients). Furthermore, plasma levels of all cholesterol parameters, TG and phospholipids decreased along with clinical disease severity as assessed by NYHA classification.

Plasma FFAs, FAs and FA composition

Patients with HF had significantly increased plasma levels of FFAs and the plasma FFA levels increased along with increasing NYHA class (Table S1). There were also several significant alterations in FA composition in patients with HF as compared with healthy controls (Tables 2 and 3 and Table S2). First, patients with HF were characterized by decreased weight per cent (wt.%) of n-3 and n-6 PUFAs. Second, the decrease in n-3 PUFAs was primary reflected in a decreased content of C20:5n-3 [eicosapentaenoic acid (EPA)], C20:4n-3, C22:5n-3 and C22:6n-3 [docosahexaenoic acid (DHA)]. Third, the decrease in wt.% of n-6 PUFAs in HF was primarily reflected in a decrease in the content of C18:2n-6 (linoleic acid). In fact, patients with HF had significant increased proportions of C22:5n-6 and C22:4n-6, and no changes in C20:4n-6. Fourth, in contrast to the decrease in PUFAs, patients with HF had an increased wt.% of monounsaturated FAs (MUFAs), primarily reflected in an increased proportion of C16:1n-9 (palmitoleic acid), C16:1n-7, C18:1n-9 (oleic acid), C18:1n-7 (vaccenic acid) and C20:1n-9. Finally, the wt.% of several of the long saturated FAs was significantly decreased in patients with HF (i.e., C22:0, C23:0 and C24:0), potentially reflecting increased oxidation of these readily oxidized FAs.

Table 2.   Fatty acid (FA) composition in patients with heart failure (HF) and controls
 ControlHFHF, non-statin users
(n = 44)(n = 183)(n = 106)
  1. Results are presented as mean ± SD. *P < 0.05; **P < 0.001 vs. controls.

  2. wt.%, weight per cent; PUFA, polyunsaturated FA.

Total FAs (μg/mL plasma)3674 ± 9613461 ± 1137*3594 ± 1182
wt.% of total FAs
 Saturated FAs32.72 ± 1.9233.29 ± 1.7633.17 ± 1.62
 Monounsaturated FAs22.48 ± 2.4526.38 ± 3.92 **26.14 ± 3.61**
 n-3 PUFA7.66 ± 2.326.16 ± 2.03**5.79 ± 1.77**
 n-6 PUFA36.71 ± 3.2533.74 ± 4.74**34.47 ± 4.36*
 n-3/n-6 PUFA0.21 ± 0.070.19 ± 0.14*0.17 ± 0.06**
Table 3.   Fatty acids (FAs) with altered levels (wt.% of total FAs) in patients with heart failure (HF) compared with healthy controls
FAsaControl (n = 44)HF (n = 183)HF, non-statin users (n = 106)
  1. Results are presented as mean ± SD. *P < 0.05; **P < 0.01; ***P < 0.001 vs. controls.

  2. wt.%, weight per cent; aCarbon (C) atoms: double bonds, number (n) of carbons between the double bond and the omega carbon; C16:1n-9, palmitoleic acid; C18:2n-6, linoleic acid; C18:1n-9, oleic acid; C18:1n-7, vaccenic acid; C20:5n-3, eicosapentaenoic acid; C20:4n-3, eicosatetraenoic acid. A complete list including FAs not altered in HF is shown in Table S2.

C10:00.04 ± 0.070.02 ± 0.01***0.02 ± 0.02***
C14:01.00 ± 0.400.92 ± 0.27*0.93 ± 0.31
C15:00.20 ± 0.070.19 ± 0.04*0.18 ± 0.04*
C16:1n-90.20 ± 0.070.25 ± 0.14***0.24 ± 0.10***
C16:1n-71.40 ± 0.401.88 ± 0.68***1.88 ± 0.70***
C16:021.72 ± 1.4623.05 ± 1.89***22.93 ± 1.61***
C18:2n-628.71 ± 3.0525.58 ± 4.33***26.52 ± 4.21**
C18:1n-917.87 ± 2.2521.06 ± 3.52***20.88 ± 3.07***
C18:1n-71.44 ± 0.201.72 ± 0.27***1.70 ± 0.28***
C20:5n-32.25 ± 1.261.48 ± 0.95***1.33 ± 0.78***
C20:4n-30.16 ± 0.070.12 ± 0.05***0.12 ± 0.04***
C20:3n-90.10 ± 0.070.12 ± 0.05*0.11 ± 0.08
C20:1n-90.15 ± 0.070.16 ± 0.05*0.15 ± 0.04
C20:1n-70.011 ± 0.0070.014 ± 0.001***0.014 ± 0.004***
C20:00.20 ± 0.0270.19 ± 0.04**0.18 ± 0.03**
C21:5n-30.015 ± 0.0130.010 ± 0.010*0.009 ± 0.007*
C22:6n-33.73 ± 0.993.24 ± 1.08**3.08 ± 0.96***
C22:5n-60.07 ± 0.020.09 ± 0.03***0.09 ± 0.03***
C22:4n-60.13 ± 0.070.15 ± 0.04***0.15 ± 0.04***
C22:5n-30.74 ± 0.130.61 ± 0.14***0.59 ± 0.13***
C22:00.58 ± 0.070.45 ± 0.14***0.46 ± 0.10***
C23:00.26 ± 0.070.19 ± 0.05***0.19 ± 0.05***
C24:1n-91.03 ± 0.200.92 ± 0.27*0.92 ± 0.28*
C24:00.55 ± 0.070.37 ± 0.14***0.38 ± 0.09***

Plasma FFAs, FAs and FA composition in relation to use of statins or beta-adrenergic receptor blockers and aetiology of HF

In contrast to the effects on lipid parameters, the use of statins had no influence on plasma FFA levels and only a minor influence on FA composition. For most FAs, there was no difference between statin users and non-statin users (data not shown). However, statin users had a lower wt.% of C18:2n-6 (linoleic acid; 24.40 ± 8.02 vs. 26.52 ± 4.21, P = 0.006) and a higher wt.% of C20:5n-3 (EPA; 1.66 ± 0.97 vs. 1.33 ± 0.78, P = 0.008) and n-3 PUFAs (6.61 ± 0.24 vs. 5.79 ± 1.77, P = 0.002) than non-statin users, resulting in a higher n-3/n-6 ratio in patients with HF who were receiving statins (0.21 ± 0.01 vs. 0.17 ± 0.06, P < 0.001). Moreover, these differences between patients with HF and healthy controls were also seen for most of the FAs when only including non-statin users in the HF group (Tables 2 and 3).

As beta-adrenergic activation may potentially influence lipolysis, we investigated whether beta-adrenergic receptor blocker treatment had any impact on FA composition; we did not find any influence of treatment (data not shown).

Twenty-two HF patients with CAD and 72 HF patients with DCM were not receiving treatment with statins. When comparing these two subgroups (i.e., patients with CAD and patients with DCM who were not receiving statin treatment), we observed no significant difference in FFA levels. However, several of the FAs that were significantly altered in patients with HF compared with controls were altered to an even greater extent in those with CAD compared with those with DCM. Wt.% of C16:1n-9, C18:1n-9 and C18:1n-7 were significantly higher and wt.% of C18:2n-6, C22:0, C23:0 and C24:0 significantly lower in HF patients with CAD compared with those with DCM (P < 0.05 for all comparisons). In addition, levels of saturated FAs and MUFAs were higher and n-6 PUFAs lower in HF patients with CAD compared with those with DCM (P < 0.05). However, except for C18:2n-6, those FAs significantly altered in plasma in the HF group compared with controls were also significantly altered in the DCM subgroup.

Low levels of C20:4n-3 (eicosatetraenoic acid) and high levels of C18:1n-7 (vaccenic acid) are associated with mortality in HF patients

During a mean follow-up of 24 ± 12 months, 46 patients died (n = 33) or underwent heart transplantation (n = 13; i.e., anticipated mortality). When including FA parameters that were significantly changed in the HF population compared with healthy controls, ROC analysis identified several markers that were significantly associated with mortality in these patients (Table 4). Thus, univariate analyses showed that high wt.% of C18:1n-9 (oleic acid; P = 0.028) and low wt.% of C22:6n-3 (P = 0.021), C23:0 (P = 0.037) and C24:0 (P = 0.030) were significantly associated with total mortality. However, the strongest associations with mortality were observed for low wt.% of 20:4n-3 (eicosatetraenoic acid) and high wt.% of C18:1n-7 (vaccenic acid) (Table 4). Indeed, after adjustment in multivariable analysis for type 2 diabetes, hypertension, age, statin use, creatinine, Nt-proBNP and hsCRP, these parameters remained significant predictors of total mortality in this HF population (Table 4). Kaplan–Meier curves showed a higher mortality in the lowest 20:4n-3 tertile and in particular in the highest C18:1n-7 tertile as compared with the other tertiles (Fig. 1).

Table 4.   Associations between free fatty acids (FFAs), fatty acid (FA) composition (wt.% of total FAs) and mortality measures (composite of death and HTx) in patients with heart failure (HF) (n = 183)
 ROCCox model 1Cox model 2
AUC (95% CI), PRR (95% CI), PRR (95% CI), P
  1. wt.%, weight per cent; HTx, heart transplantation; MUFA, monounsaturated FA; ROC, receiver operating characteristics; AUC, area under the curve. AUC > 0.5 indicates a positive association with increased mortality whereas AUC < 0.5 indicates a negative association. Model 1 shows the unadjusted association between FFAs and FA composition and mortality measures, whereas model 2 is adjusted for type 2 diabetes, hypertension, age, statin use, creatinine, Nt-proBNP and hsCRP, variables known to influence the prognosis in patients with HF.

FFAs0.61 (0.51–0.70), 0.0341.39 (0.96–2.01), 0.0811.28 (0.88–1.86), 0.199
FAs
 MUFAs0.64 (0.54–0.73), 0.0071.38 (0.96–2.00), 0.0831.16 (0.79–1.71), 0.457
 C18:1n-90.63 (0.54–0.73), 0.0081.52 (1.05–2.21), 0.0281.27 (0.86–1.85), 0.227
 C18:1n-70.66 (0.58–0.75), 0.0011.81 (1.24–2.66), 0.0021.55 (1.03–2.34), 0.037
 C20:5n-30.40 (0.30–0.49), 0.0390.76 (0.53–1.08), 0.1260.70 (0.48–1.04), 0.074
 C20:4n-30.30 (0.21–0.39), <0.0010.57 (0.38–0.84), 0.0040.56 (0.37–0.87), 0.009
 C22:00.38 (0.28–0.48), 0.0150.71 (0.49–1.03), 0.0730.70 (0.61–1.39), 0.699
 C22:6n-30.44 (0.35–0.53), 0.1930.51 (0.29–0.90), 0.0210.58 (0.31–1.10), 0.094
 C23:00.38 (0.28–0.48), 0.0200.67 (0.46–0.98), 0.0370.82 (0.54–1.24), 0.347
 C24:00.36 (0.26–0.46), 0.0070.66 (0.45–0.96), 0.0300.90 (0.59–1.38), 0.628
image

Figure 1.  Kaplan–Meier curves showing associations between tertiles of C20:4n-3 and C18:1n-7. Kaplan–Meier curves showing associations between tertiles (T) of C20:4n-3 (eicosatetraenoic acid) (a) and C18:1n-7 (vaccenic acid) (b), given as wt.% of total FAs, and total mortality in HF patients. During a mean follow-up of 24 ± 12 months, 46 of 183 patients died or underwent heart transplantation (i.e., total mortality). The molecular structure of the two FAs is shown under each panel. ω indicates the omega carbon.

Download figure to PowerPoint

Correlations between FFAs, FAs and FA composition and markers of myocardial function and inflammation

Analysis of the correlations between FFAs and FA parameters identified as significantly associated with mortality in Table 4and NYHA functional class, markers of myocardial function (CI and Nt-proBNP) and inflammation (hsCRP) in patients with HF revealed several significant findings (Table 5). First, high plasma levels of FFAs were correlated with increasing NYHA class, increased Nt-proBNP levels and raised hsCRP levels. Second, the proportion of MUFAs was positively correlated with NYHA class and Nt-proBNP levels and negatively correlated with CI. In line with this, a high proportion of C18:1n-9 (oleic acid) and C18:1n-7 (vaccenic acid) was positively correlated with hsCRP level, NYHA class and impaired myocardial function as assessed by a positive correlation with Nt-proBNP and a negative correlation with CI. Fourth, the decrease in n-3 PUFAs was correlated with raised hsCRP level [C20:5n-3 (EPA) and C20:4n-3 (eicosatetraenoic acid)], increased NYHA class (C20:5n-3) and increased Nt-proBNP level [C20:5n-3 and C22:6n-3 (DHA)]. Finally, the decrease in the wt.% of the long saturated FAs C22:0, C23:0 and C24:0 was significantly correlated with raised hsCRP levels, increasing NYHA class and impaired myocardial function as assessed by a negative correlation with Nt-proBNP and a positive correlation with CI. In general, several of the correlations were weak, suggesting that these FAs may reflect mechanisms that are only partly accounted for by the clinical variables Nt-proBNP levels and hsCRP levels.

Table 5.   Correlation between free fatty acids (FFAs), fatty acid (FA) composition (wt.% of total FAs) and New York Heart Association (NYHA) functional class, cardiac index, plasma N-terminal pro-B-type natriuretic peptide (Nt-proBNP) and high-sensitivity C-reactive protein (hsCRP) in patients with heart failure (n = 183)
 NYHA classCardiac indexNt-proBNPhsCRP
R (p-value)R (p-value)R (p-value)R (p-value)
  1. wt.%, weight per cent; MUFAs, monounsaturated FAs.

FFAs0.20 (0.006)−0.05 (0.646)0.15 (0.042)0.20 (0.007)
FAs
 MUFAs0.26 (<0.001)−0.22 (0.024)0.35 (<0.001)0.17 (0.021)
 C18:1n-90.27 (<0.001)−0.20 (0.037)0.16 (0.029)0.33 (<0.001)
 C18:1n-70.27 (<0.001)−0.27 (0.005)0.26 (<0.001)0.32 (<0.001)
 C20:5n-3−0.07 (0.338)0.06 (0.543)−0.04 (0.554)−0.17 (0.017)
 C20:4n-3−0.27 (<0.001)0.16 (0.100)−0.33 (<0.001)−0.25 (0.001)
 C22:0−0.21 (0.004)0.36 (<0.001)−0.23 (0.002)−0.29 (<0.001)
 C22:4n-6−0.06 (0.495)0.08 (0.338)−0.27 (<0.001)0.01 (0.925)
 C23:0−0.24 (0.001)0.24 (0.015)−0.22 (0.003)−0.34 (<0.001)
 C24:00.25 (<0.001)0.35 (<0.001)−0.25 (<0.001)−0.24 (<0.001)

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Conflict of interest statement
  9. Acknowledgements
  10. References
  11. Supporting Information

Whereas several studies have shown an association between FA composition and CAD, stroke and arrhythmias, evidence for an association between FA composition and HF is limited. Results from a few studies have suggested that increased intake of fish and/or n-3 PUFAs (omega-3 FAs) is associated with reduced risk of HF [10, 11]. Moreover, Yamagishi et al. [6] showed that higher levels of saturated FAs and lower levels of n-6 (20:3n-6) and n-3 (22:6n-3) PUFAs were associated with incident HF. Furthermore, the GISSI HF investigators reported that supplementation with n-3 PUFAs provided an advantage in terms of mortality and admission to hospital for cardiovascular reasons in patients with HF [8]. In the present study, we extend these findings by showing several significant changes in the composition of FAs in patients with HF, with decreased levels of n-3 and n-6 PUFAs, increased levels of MUFAs and decreased levels of several readily oxidized long-chain saturated FAs as the most prominent findings. These changes in FAs were significantly associated with an increase in NYHA class, impaired cardiac function (CI and Nt-proBNP) and systemic inflammation (hsCRP). Moreover, high levels of C18:1n-7 (MUFA) and low levels of C20:4n-3 (n-3 PUFA) were significantly associated with total mortality in this HF population, even after adjustment for several confounders. Although our data should be interpreted with caution, our findings may further support a link between disturbed FA composition and the progression of HF, and may suggest that some of these FAs could be further assessed as prognostic markers in patients with this disease. It is possible that some specific biological properties of C18:1n-7 and 20:4n-3 may overcome their low proportions in plasma and the heart, but this remains to be determined in future studies.

Decreased levels of n-3 PUFAs in HF

The results of several studies suggest that n-3 PUFAs may exert anti-inflammatory and platelet-inhibiting effects [12, 13]. Also, epidemiological and experimental studies have shown that these PUFAs may reduce vascular resistance, attenuate the vasoconstrictor response to angiotensin II and improve left ventricular diastolic function [14–16]. Moreover, it has also been suggested that n-3 PUFAs have anti-arrhythmic activity [17], and the reduction in total mortality during n-3 PUFA supplementation in patients with HF has been shown, at least partly, to reflect a reduction in death caused by arrhythmias [8]. In the present study, we showed decreased levels of a number of n-3 PUFAs in patients with HF, significantly associated with enhanced inflammation and impaired myocardial function, further supporting a role for decreased n-3 PUFA levels in the progression of HF. However, whereas numerous studies have focused on C20:5n-3 (EPA) and C22:6n-3 (DHA), our findings suggest that other n-3 PUFAs may also be of importance. Indeed, whereas decreased wt.% of C20:5n-3, C20:4n-3 and C22:5n-3 was correlated with raised hsCRP levels, only C20:4n-3 (eicosatetraenoic acid) was inversely correlated with Nt-proBNP as a marker of impaired myocardial function. Also, decreased wt.% of C20:4n-3, but not of the other n-3 PUFAs, was significantly associated with total mortality, even after adjustment for several confounders.

At present, there is no firm evidence for a possible role of eicosatetraenoic acid during myocardial remodelling. However, it is tempting to hypothesize that eicosatetraenoic acid, as a mitochondrial targeted FA, could improve energy and redox status within the myocardium. Moreover, as a precursor of EPA, it is possible that eicosatetraenoic acid could more accurately reflect n-3 PUFA status than EPA. Furthermore, like EPA and other long-chain n-3 PUFAs, C20:4n-3 has been shown to have anti-inflammatory effects by inhibiting the conversion of arachidonic acid to prostaglandins [18]. Similarly, like other n-3 PUFAs, C20:4n-3 is readily incorporated into various phospholipid species, which is consistent with the possible modulatory effects on eicosanoid metabolism. Nonetheless, although further mechanistic studies are needed for confirmation, our findings suggest that the beneficial effect of n-3 PUFAs in HF and other cardiovascular disorders may not necessarily be related to EPA or any other specific PUFA, but rather involve the complex effects of several n-3 PUFAs, including eicosatetraenoic acid.

Increased levels of MUFAs in HF

In contrast to the decrease in n-3 and n-6 PUFAs, patients with HF had an increased proportion of MUFAs. Recently, Yamagishi et al. [6] showed that increased C16:1n-7 was associated with incident HF in apparently healthy individuals. Here, we extend these findings by showing an increased proportion of several MUFAs in patients with HF [i.e., C16:1n-9 (palmitoleic acid), C16:1n-7, C18:1n-9 (oleic acid), C18:1n-7 (vaccenic acid) and C20:1n-9]. Some of these (i.e., C18:1n-9, C18:1n-7 and C20:1n-9) were correlated with enhanced inflammation and impaired myocardial function. Additionally, a high wt.% of C18:1n-7 was associated with total mortality, even after adjustment for several confounders.

There is some evidence that MUFAs may have beneficial effects in various cardiovascular disorders. Thus, the substitution of MUFAs for saturated FAs in the prevention of cardiovascular disease is recommended on the basis of the benefits of the traditional Mediterranean diet [19, 20]. We found an increased proportion of C16:1n-7 (palmitoleic acid) in patients with HF, and it is interesting that this FA has recently been identified as an adipose tissue-derived lipid hormone that strongly stimulates the action of muscle insulin and suppresses hepatosteatosis [21]. However, the role of MUFAs in cardiovascular disorders is far from clear. Whereas C18:1n-7 may have some protective effects at normal physiological concentrations, it is possible that higher levels may adversely affect biologically active proteins in myocardial membranes, potentially contributing to the progression of HF. Thus, enhanced myocardial expression of C18:1n-7, an elongated product of palmitoleic acid, has been reported in hypertrophic cardiomyopathy in mice with carnitine deficiency [22], possibly promoting harmful effects. Moreover, Lemaitre et al. [23] recently found that high levels of C18:1n-7 were associated with sudden cardiac death, independent of low levels of EPA and DHA. The authors suggested that this may be because of a possible adverse effect of lysophospholipids with remaining vaccenic acid in sn1 position (after long-chain PUFA as EPA or DHA has been released from the sn2 position by phospholipase A2 during ischaemia). Also, an increased proportion of MUFAs has been reported in experimental steatohepatitis [24], and we have previously shown that in contrast to n-3 PUFAs, MUFA supplementation in rats has no effect on plasma levels of FFAs and increases plasma levels of TG [25]. Furthermore, Morel et al. [26] reported that a decrease in the tolerance of cardiomyocytes to ischaemic insults, resulting in an increased frequency of arrhythmia, was associated with alteration of the plasma lipid profile, primarily reflecting an increase in MUFAs and a decrease in PUFAs. Nonetheless, the role of MUFAs in the development of myocardial failure is at present not clear.

Decreased plasma levels of lipid parameters and increased levels of FFAs in HF

In line with previous findings, we found decreased levels of several lipid parameters and increased FFA levels in patients with HF, which were significantly correlated with disease severity. FFAs have been shown to induce increased oxygen consumption within the myocardium, despite a lack of change of mechanical activity, reflecting oxygen wasting [27]. The results of more recent studies suggest that this FFA-mediated effect may involve increased cardiac mitochondrial uncoupling proteins, which, in turn, are associated with decreased mitochondrial respiratory coupling and low cardiac efficiency [28]. However, in contrast to C18:1n-7 and C20:4n-3, we found no association between FFAs and mortality, suggesting that FA composition and certain FAs may be of more importance for the progression of HF than FFAs.

Study limitations

The present study has some limitations. A large proportion of the patients with HF included were hospitalized for evaluation of heart transplantation. As the age limit for transplantation in Norway is 65–70 years, most of the patients included were <65 years of age. Because DCM is more common in younger patients with HF, as many as 51% of our patients with HF had DCM. Thus, our data may not necessarily be representative of the situation in an older HF population. In addition, although there was no evidence of an excess intake of vitamins, fish oils, antioxidants or micronutrients, we cannot exclude the possibility that minor differences in intake of FAs could to some degree have influenced our data. Furthermore, although we used a strict and standardized blood sampling protocol with similar conditions for all patients and controls, the use of nonfasting samples is a limitation of the study. A wide range of parameters was analysed, and caution is needed when interpreting data from multiple comparisons. On the other hand, the different FAs interact within a complex network, and the presentation of several FA parameters could also be regarded as a strength of the present study. Our study was cross-sectional, and longitudinal data are needed to further elucidate the role of FAs in HF. Finally, statistical associations do not necessarily imply any causal relationships, and further mechanistic studies are required to clarify the role of FAs in the progression of HF.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Conflict of interest statement
  9. Acknowledgements
  10. References
  11. Supporting Information

In the present study, we have shown that patients with HF are characterized by a certain FA phenotype, i.e., a low proportion of some n-3 and n-6 PUFAs and a high proportion of certain MUFAs, which was significantly associated with inflammation and the degree of myocardial dysfunction. Some of these FAs (C18:1n-7 and C20:4n-3) were also associated with total mortality during follow-up. Although further studies are needed, our findings may suggest an association between plasma FA composition and the progression of HF.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Conflict of interest statement
  9. Acknowledgements
  10. References
  11. Supporting Information

This work was supported by grants from the Norwegian Council of Cardiovascular Research, the Research Council of Norway, the Southern and Eastern Norwegian regional health authority (Helse Sør-Øst), Oslo University Hospital Rikshospitalet, the Western Norwegian regional health authority (Helse Vest), the Nordic Centre of Excellence (MitoHealth) and the EU Project AtheroRemo.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Conflict of interest statement
  9. Acknowledgements
  10. References
  11. Supporting Information
  • 1
    Ventura-Clapier R, Garnier A, Veksler V. Energy metabolism in heart failure. J Physiol 2004; 555: 113.
  • 2
    Stanley WC, Recchia FA, Lopaschuk GD. Myocardial substrate metabolism in the normal and failing heart. Physiol Rev 2005; 85: 1093129.
  • 3
    Opie LH, Knuuti J. The adrenergic-fatty acid load in heart failure. J Am Coll Cardiol 2009; 54: 163746.
  • 4
    Neubauer S. The failing heart-an engine out of fuel. N Engl J Med 2007; 356: 114051.
  • 5
    Nasa Y, Sakamoto Y, Sanbe A, Sasaki H, Yamaguchi F, Takeo S. Changes in fatty acid compositions of myocardial lipids in rats with heart failure following myocardial infarction. Mol Cell Biochem 1997; 176: 17989.
  • 6
    Yamagishi K, Nettleton JA, Folsom AR. Plasma fatty acid composition and incident heart failure in middle-aged adults: the Atherosclerosis Risk in Communities (ARIC) Study. Am Heart J 2008; 156: 96574.
  • 7
    Yamagishi K, Iso H, Date C et al. , Japan Collaborative Cohort Study for Evaluation of Cancer Risk Study Group. Fish, omega-3 polyunsaturated fatty acids, and mortality from cardiovascular diseases in a nationwide community-based cohort of Japanese men and women the JACC (Japan Collaborative Cohort Study for Evaluation of Cancer Risk) Study. J Am Coll Cardiol 2008; 52: 98896.
  • 8
    Gissi-HF Investigators, Tavazzi L, Maggioni AP, Marchioli R et al. Effect of n-3 polyunsaturated fatty acids in patients with chronic heart failure (the GISSI-HF trial): a randomised, double-blind, placebo-controlled trial. Lancet 2008; 372: 122330.
  • 9
    Wergedahl H, Liaset B, Gudbrandsen OA et al. Fish protein hydrolysate reduces plasma total cholesterol, increases the proportion of HDL cholesterol, and lowers acyl-CoA:cholesterol acyltransferase activity in liver of Zucker rats. J Nutr 2004; 134: 13207.
  • 10
    Mozaffarian D, Bryson CL, Lemaitre RN, Burke GL, Siscovick DS. Fish intake and risk of incident heart failure. J Am Coll Cardiol 2005; 45: 201521.
  • 11
    Duda MK, O’Shea KM, Stanley WC. Ω-3 polyunsaturated fatty acid supplementation for the treatment of heart failure: mechanisms and clinical potential. Cardiovasc Res 2009; 84: 3341.
  • 12
    Dyerberg J, Bang HO, Stoffersen E, Moncada S, Vane JR. Eicosapentaenoic acid and prevention of thrombosis and atherosclerosis? Lancet 1978; 2: 1179.
  • 13
    Calder PC. n-3 polyunsaturated fatty acids, inflammation, and inflammatory diseases. Am J Clin Nutr 2006; 83(Suppl): 1505S19S.
  • 14
    Raitt MH, Connor WE, Morris C et al. Fish oil supplementation and risk of ventricular tachycardia and ventricular fibrillation in patients with implantable defibrillators: a randomized controlled trial. JAMA 2005; 293: 288491.
  • 15
    Chin JP, Gust AP, Nestel PJ, Dart AM. Marine oils dose-dependently inhibit vasoconstriction of forearm resistance vessels in humans. Hypertension 1993; 21: 228.
  • 16
    Li Q, Zhang Q, Wang M et al. Docosahexaenoic acid affects endothelial nitric oxide synthase in caveolae. Arch Biochem Biophys 2007; 466: 2509.
  • 17
    London B, Albert C, Anderson ME et al. Omega-3 fatty acids and cardiac arrhythmias: prior studies and recommendations for future research: a report from the National Heart, Lung, and Blood Institute and Office of Dietary Supplements Omega-3 Fatty Acids and their Role in Cardiac Arrhythmogenesis Workshop. Circulation 2007; 116: e32035.
  • 18
    Ghioni C, Porter AEA, Taylor GW, Tocher DR. Metabolism of 18:4n-3 (stearidonic acid) and 20:4n-3 in salmonid cells in culture and inhibition of the production of prostaglandin F (PGF) from 20:4n-6 (arachidonic acid). Fish Physiol Biochem 2002; 27: 8196.
  • 19
    Smith RD, Kelly CN, Fielding BA et al. Long-term monounsaturated fatty acid diets reduce platelet aggregation in healthy young subjects. Br J Nutr 2003; 90: 597606.
  • 20
    Appel LJ, Sacks FM, Carey VJ et al. , OmniHeart Collaborative Research Group. Effects of protein, monounsaturated fat, and carbohydrate intake on blood pressure and serum lipids: results of the OmniHeart randomized trial. JAMA 2005; 294: 245564.
  • 21
    Cao H, Gerhold K, Mayers JR, Wiest MM, Watkins SM, Hotamisligil GS. Identification of a lipokine, a lipid hormone linking adipose tissue to systemic metabolism. Cell 2008; 134: 93344.
  • 22
    Takahashi R, Okumura K, Asai T et al. Dietary fish oil attenuates cardiac hypertrophy in lipotoxic cardiomyopathy due to systemic carnitine deficiency. Cardiovasc Res 2005; 68: 21323.
  • 23
    Lemaitre RN, King IB, Sotoodehnia N et al. Endogenous red blood cell membrane fatty acids and sudden cardiac arrest. Metabolism 2010; 59: 102934.
  • 24
    Li ZZ, Berk M, McIntyre TM, Feldstein AE. Hepatic lipid partitioning and liver damage in nonalcoholic fatty liver disease: role of stearoyl-CoA desaturase. J Biol Chem 2009; 284: 563744.
  • 25
    Halvorsen B, Rustan AC, Madsen L et al. Effects of long-chain monounsaturated and n-3 fatty acids on fatty acid oxidation and lipid composition in rats. Ann Nutr Metab 2001; 45: 307.
  • 26
    Morel S, Berthonneche C, Tanguy S et al. Insulin resistance modifies plasma fatty acid distribution and decreases cardiac tolerance to in vivo ischaemia/reperfusion in rats. Clin Exp Pharmacol Physiol 2003; 30: 44651.
  • 27
    Mjøs OD. Effect of free fatty acids on myocardial function and oxygen consumption in intact dogs. J Clin Invest 1971; 50: 13869.
  • 28
    Murray AJ, Cole MA, Lygate CA et al. Increased mitochondrial uncoupling proteins, respiratory uncoupling and decreased efficiency in the chronically infarcted rat heart. J Mol Cell Cardiol 2008; 44: 694700.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Conflict of interest statement
  9. Acknowledgements
  10. References
  11. Supporting Information

Table S1. Plasma levels of lipids and free fatty acids (FFAs) in patients with heart failure (HF) and healthy controls.

Table S2. Fatty acid (FA) composition (wt.% of total FAs) in patients with heart failure (HF) and healthy controls.

FilenameFormatSizeDescription
JOIM_2384_sm_Supplemental-Tables.doc94KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.