Abstract
Aim: Malnutrition has been identified to be an independent predictor of morbidity and mortality in patients with chronic heart failure (CHF). However, the pathophysiological mechanisms underlying this pathway remain unclear.
Methods: Nutritional screening was performed using the controlling nutritional status (CONUT) score, which was calculated using the serum albumin and total cholesterol levels and lymphocyte number, in 114 CHF patients with a mean left ventricular ejection fraction of 26.6% ± 6.4%. The carotid intima-media thickness (CIMT) is correlated with carotid atherosclerosis and is a significant predictor of future cardiovascular events. Peripheral blood mononuclear cells (PBMCs) were isolated, and the production of monocyte tumor necrosis factor (TNF)-α was measured and expressed as mean ± SD (pg/mL/106 PBMCs).
Results: A multivariate linear regression analysis showed that the production of monocyte TNF-α (β coefficient = 0.434, p < 0.001) and mean CIMT (β coefficient = 0.204, p = 0.006) were independent determinants of the CONUT score. During a median follow-up of 67.5 months, 45 patients experienced cardiac events, including 16 cardiac deaths and 29 readmissions for worsening CHF. A multivariate Cox hazard analysis demonstrated that a monocyte TNF-α level of ≥ 4.1 pg/mL/106 PBMCs (hazard ratio (HR), 14.10; 95% confidence interval (CI), 2.55–77.92; p = 0.002) and CONUT score of ≥ 3 (HR, 11.97; 95% CI, 2.21–64.67; p = 0.004) were independently associated with the incidence of cardiac events.
Conclusions: These data indicate that a poor nutritional status as assessed using the CONUT score and atherosclerosis as indicated by CIMT is significantly associated with inflammation and predicts poor outcomes in patients with CHF.
Keywords: Malnutrition, Inflammation, Chronic heart failure
Introduction
Malnutrition has been identified to be an independent risk factor of morbidity and mortality in patients with chronic heart failure (CHF)1, 2). Nutritional intervention may prevent complications and increase the quality of life in CHF patients3, 4). Therefore, it is very important to evaluate the nutritional status of CHF patients. The controlling nutritional status (CONUT) score was developed by Ignacio de Ulíbari et al.5) as a screening tool for identifying under-nutrition in hospital populations. This score is based on three parameters, the serum albumin and total cholesterol levels and the total lymphocyte count. Therefore, this score enables evaluation of the protein reserve, calorie depletion, and immune defenses.
Proinflammatory cytokines, such as tumor necrosis factor (TNF)-α, have been implicated in the pathogenesis and progression of heart failure as mediators of myocardial dysfunction and remodeling6–8). The plasma levels of TNF-α have been reported to be significantly increased in CHF patients with malnutrition9), and it has been postulated that the common link between CHF and malnutrition may be inflammation10).
Arterial wall thickening can be assessed in vivo according to ultrasound measurements of the carotid intima-media thickness (CIMT). The CIMT is correlated with coronary and carotid atherosclerosis and is a significant predictor of future cardiovascular events. Thus, CIMT has been widely used in many clinical studies as a surrogate marker of coronary and carotid atherosclerosis and the risk of cardiovascular diseases11–15).
Nevertheless, no previous studies have examined the relationships between the CONUT score, production of monocyte TNF-α, and the long-term prognosis in patients with CHF.
Methods
Patient Population
This study was a prospective and observational study that included a total of 114 selected (nonconsecutive) eligible patients with CHF (85 men and 29 women; mean age, 66.0 ± 11.3 years), with a mean left ventricular ejection fraction (LVEF) of 26.6% ± 6.4%. All patients were recruited from an outpatient clinic at either the Nippon Medical School Hospital or Tama-Nagayama Hospital between January 2000 and December 2011.
Patients were enrolled if they had dyspnea or fatigue at rest or on minimal exertion [New York Heart Association (NYHA) class II or III] and a LVEF of ≤ 45% due to ischemic or dilated cardiomyopathy. The etiology of CHF was dilated cardiomyopathy (DCM) in 88 patients and ischemic cardiomyopathy (ICM) in 26 patients. DCM was defined as a normal coronary arteriogram together with severe hypokinesis of the left ventricular wall motion as determined on left ventriculography and according to typical pathological findings of an endomyocardial biopsy of the left ventricle. ICM was defined as a history of myocardial infarction with significant coronary artery disease (> 70% luminal stenosis in at least two major coronary arteries). All patients with ICM received coronary artery bypass grafting (CABG) or percutaneous coronary intervention (PCI) before enrollment, and there was no evidence of myocardial ischemia using a treadmill exercise test or exercise perfusion myocardial scintigraphy. In addition, all eligible patients were stabilized with respect to CHF using standard medical treatment, including angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor II blockers (ARBs), β-blockers, furosemide, spironolactone, or any combination of these medications for at least 3 years prior to enrollment in this study. No patients received any additional treatment for CHF during the follow-up period.
Exclusion criteria were as follows. All patients with clinical signs of acute infection, autoimmune disorders, severe renal (serum creatinine level of > 2.0 mg/dL) or hepatic disease, and those with suspected malignancy were excluded from the present study.
In addition, all patients in the acute decompensated stage of CHF and those with NYHA class IV were excluded from this study. There were no patients with a history of hypertensive heart disease, valvular heart disease, or ischemic heart disease with multivessel coronary disease without myocardial infarction.
LVEF was measured by a blinded trained sonographer using echocardiography with a LOGIQ 7 system (GE; Milwaukee, WI) within 1 week of measuring the biochemical markers. The patients were placed in the left lateral recumbent position, and LVEF was calculated according to the Simpson method using apical four- and two-chamber views. All echocardiograms of the eligible patients were measured and analyzed by the blinded trained cardiologist.
The investigation protocol was designed according to the guidelines of the institutional ethics committee. All selected (nonconsecutive) eligible patients provided their written informed consent to participate in the study. None of the patients dropped out from this study during the follow-up period.
Reagents and Laboratory Measurements
Hank's balanced salt solution was obtained from Sigma (Tokyo, Japan) and RPMI 1640 medium was obtained from GIBCO (Tokyo, Japan). The endotoxin- tested solution, Lymphoprep, was obtained from Nycomed Pharma AS (Oslo, Norway). All media, including RPMI and Hank's balanced salt solution, were tested for endotoxin, which was present at a level of < 5 pg per mL in each medium.
Citrated venous peripheral blood was collected from CHF patients. Peripheral blood mononuclear cells (PBMCs: monocytes + lymphocytes) were isolated as previously described16, 17). Preliminary studies of 10 patients with CHF showed that the production of cytokines by PBMCs was only slightly higher than that by purified monocytes (TNF-α: 3.9 ± 1.6 vs. 3.8 ± 1.5 pg/mL/106 PBMCs, p = 0.016). These data indicated that monocytes are the principle cells capable of cytokine synthesis and that lymphocytes may contribute, at least in part, to the production of cytokines by PBMCs. However, the difference between PBMCs and monocytes was very small; therefore, PBMCs were routinely cultured for this study, as previously described.
The production of TNF-α by PBMCs was apparent after 4 h, and the peak concentration was achieved between 18 and 24 h of culture, after which it declined slightly. Therefore, PBMCs were cultured in 96-well plates for 24 h, and the supernatants were separated from PBMCs via centrifugation and then stored at −80°C until the cytokine levels were determined. The production of TNF-α by PBMCs was measured using a specific enzyme-linked immunosorbent assay (ELISA) with a commercially available system (R&D Systems), and the results are expressed as mean ± SD (pg/mL/106 PBMCs). The intra- and inter-assay coefficients of variation were < 5% for all ELISAs, and all samples were analyzed in duplicate.
The level of C-reactive protein (CRP) was measured using an immunoturbidimetry assay. The plasma B-type natriuretic peptide (BNP) concentration was determined with a specific immunoradiometric assay for human BNP using commercial kits (Shionoria Kit, Shionogi and Kyowa Medex, Tokyo, Japan). The performance characteristics of the Shionoria BNP kit include a coefficient of variance of 2.5%–4.3% (n = 10) and an analytical range of 4-2,000 pg/mL.
To accurately evaluate the renal function, the estimated glomerular filtration rate (eGFR) was calculated according to the following equation for Japanese subjects, as recommended by the Japanese Society of Nephology18): eGFR (mL/min/1.73 m2) = 194 × serum creatinine−1.094 (mg/dL) × age−0.287 (years) × 0.739, if female.
CIMT
CIMT was determined for all patients on the basis of entry in this study. The right and left carotid arteries were examined, and mean CIMT (mCIMT) was estimated according to a standardized protocol by a blinded trained sonographer using B-mode ultrasound with a 10 MHz linear probe (LOGIQ 7, GE, Milwaukee, WI). CIMT was measured at three points in each common carotid artery 10 mm proximal to the site of bifurcation, and the mean value of six measurements for the right and left carotid arteries was calculated and used for the subsequent analysis. The intra- and inter-observer coefficients of variation for the repeated measurements of CIMT in our laboratory were 2.9% and 3.2%, respectively.
Evaluation of the Nutritional Status
In this study, the CONUT score was used to evaluate the nutritional status in patients with CHF. This system, developed for hospitalized patients, uses the following three parameters: the serum albumin level (g/dL), total cholesterol level (mg/dL), and lymphocyte count (count/mL)5). It thus enables evaluations of the protein reserves, calorie depletion, and immune defenses, respectively (Table 1).
Table 1. Assessment of the CONUT score.
| Parameters | Score | |||
|---|---|---|---|---|
| Serum albumin (g/dL) | ≥ 3.5 | 3.0–3.49 | 2.50–2.99 | < 2.5 |
| Albumin score | 0 | 2 | 4 | 6 |
| Total cholesterol (mg/dL) | ≥ 180 | 140–179 | 100–139 | < 100 |
| Cholesterol score | 0 | 1 | 2 | 3 |
| Lymphocytes (count/mL) | ≥ 1,600 | 1,200–1,599 | 800–1,199 | < 800 |
| Lymphocyte score | 0 | 1 | 2 | 3 |
Follow-Up and Determination of Outcomes
Outcome data were collected via serial direct contact with the patients (every 4–8 weeks) at the outpatient clinic of either Nippon Medical School Hospital or Tama-Nagayama Hospital, Nippon Medical School until December 2014. All patients were followed up for a median of 67.5 months (range: 12–180 months) to determine the incidence of cardiac events, such as cardiac death and readmission due to worsening CHF. None of the patients dropped out from this study.
Statistical Analysis
The results are presented as mean ± SD for continuous variables and the percentage of the total number of patients for categorical variables. Student's t-test for independent samples and the chi-square test were used for comparisons of continuous and categorical variables, respectively. The cytokine, triglyceride and CRP values, and the CONUT score exhibited skewed distributions; therefore, the Mann–Whitney U test was used for unpaired comparisons between groups and Wilcoxon's signed-rank test was used for paired comparisons within groups, with the data expressed as median (25th–75th percentile). Bivariate correlations between parameters were assessed with the Pearson or Spearman correlation (r) coefficient for normal or skewed distributions, respectively. Comparisons among three groups were made with a two-way analysis of variance followed by the Bonferroni correction.
The associations between the CONUT score and mCIMT and other variables, were explored using multiple linear regression analyses with forward stepwise selection of covariates. A receiver operating characteristic (ROC) curve analysis was performed to determine the optimal cut-off values for age, the CONUT score, serum albumin levels, body mass index (BMI), mCIMT, systolic blood pressure, heart rate, hemoglobin (Hb), fasting plasma glucose (FPG), eGFR, LVEF, monocyte TNF-α production, and plasma CRP and BNP levels as predictors of cardiac events.
The event-free survival curves were calculated using the Kaplan–Meier method, and differences between the curves were evaluated using the log-rank test. Univariate and multivariate Cox regression analyses were employed to calculate the estimated hazard ratio (HR) with 95% confidence interval (CI), where appropriate. The variables were entered into a multivariate model for factors with a p value of ≤ 0.05 in the univariate analysis. The examined variables included patient age; BMI; mCIMT; LVEF; CONUT score; and the eGFR, CRP, BNP, and Hb levels as well as the production of monocyte TNF-α at baseline and the use of β-blockers or spironolactone. The Statistical Package for Social Science (SPSS) for Windows, version 22.0 (IBM, Tokyo) software package, was used for all statistical analyses. A p value of <0 .05 was considered to be statistically significant.
Results
Study Population
The median value (25th–75th percentile) of the CONUT score was 2 (0, 4). The patients were categorized as follows: CONUT 0–1 (n = 52), CONUT 2 (n = 16), and CONUT ≥ 3 (n = 46). The baseline clinical characteristics of patients categorized according to the CONUT score are shown in Table 2. Patients with a high CONUT score were older; had lower BMI, Hb, TG, low density lipoprotein (LDL) cholesterol, eGFR, and LVEF values; and higher heart rate and mCIMT values as well as CRP and BNP levels than those with a low CONUT score. In addition, the patients with a high CONUT score had a more severe New York Heart Association (NYHA) class and were less likely to have a history of dyslipidemia and more likely to have a history of hypertension than those with a low CONUT score (Table 2). However, there were no significant differences among the three groups with respect to gender; systolic blood pressure (SBP); high density lipoprotein (HDL) cholesterol; FPG and Hb A1c (HbA1c) levels; etiology of CHF (DCM or ICM); prevalence of diabetes and smoking; or the use of angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor blockers (ARBs), calcium channel blockers (CCBs), aspirin, β-blockers, spironolactone, or furosemide (Table 2).
Table 2. Clinical characteristics and blood chemical variables of the patients according to the CONUT score.
| All (n = 114) | CONUT (0–1: n = 52) | CONUT (2: n = 16) | CONUT (≥ 3: n = 46) | P value | |
|---|---|---|---|---|---|
| Age (years) | 66.0 ± 11.3 | 61.8 ± 12.6 | 69.7 ± 7.0 | 69.4 ± 9.2 | 0.001 |
| Men, n (%) | 85 (74.6) | 42 (80.8) | 12 (75.0) | 31 (67.4) | 0.316 |
| DCM/ICM | 88/26 | 41/11 | 13/3 | 34/12 | 0.774 |
| NYHA class (I/II/III/IV) | 0/68/46/0 | 0/48/4/0 | 0/10/6/0 | 0/14/32/0 | < 0.001 |
| Serum albumin (g/dL) | 3.5 ± 0.5 | 3.9 ± 0.3 | 3.4 ± 0.1 | 3.1 ± 0.3 | < 0.001 |
| Total cholesterol (mg/dL) | 204 ± 27 | 217 ± 22 | 213 ± 24 | 188 ± 23 | < 0.001 |
| Lymphocyte (count/mL) | 1781 ± 640 | 2105 ± 401 | 2127 ± 455 | 1293 ± 607 | < 0.001 |
| BMI (kg/m2) | 21.9 ± 2.2 | 23.4 ± 1.9 | 21.4 ± 2.0 | 20.3 ± 1.5 | < 0.001 |
| Systolic BP (mmHg) | 119 ± 8 | 120 ± 8 | 116 ± 7 | 117 ± 8 | 0.165 |
| Heart rate (beats/min) | 75 ± 7 | 73 ± 7 | 78 ± 6 | 77 ± 7 | 0.002 |
| Smoking (%) | 22 (19.3) | 7 (13.5) | 4 (25.0) | 11 (23.9) | 0.350 |
| Hypertension (%) | 36 (31.6) | 22 (42.3) | 2 (12.5) | 12 (26.1) | 0.047 |
| Diabetes (%) | 28 (24.6) | 13 (25.0) | 4 (25.0) | 11 (23.9) | 0.991 |
| Dyslipidemia (%) | 53 (46.5) | 25 (48.1) | 13 (81.3) | 15 (32.6) | 0.003 |
| Hemoglobin (g/dL) | 12.2 ± 1.3 | 12.8 ± 1.2 | 11.6 ± 1.0 | 11.7 ± 1.2 | < 0.001 |
| CRP (mg/dL) | 0.41 (0.32, 0.58) | 0.36 (0.24. 0.39) | 0.38 (0.35, 0.46) | 0.63 (0.48, 0.69) | < 0.001 |
| LDL-choresterol (mg/dL) | 130 ± 23 | 140 ± 20 | 137 ± 22 | 117 ± 19 | < 0.001 |
| HDL-choresterol (mg/dL) | 45 ± 6 | 46 ± 6 | 44 ± 8 | 45 ± 5 | 0.289 |
| Triglycerides (mg/dL) | 140 (120, 152) | 142 (128, 152) | 149 (140, 171) | 122 (102, 140) | < 0.001 |
| FPG (mg/dL) | 106 ± 16 | 104 ± 20 | 106 ± 14 | 109 ± 17 | 0.280 |
| HbAlc (JDS; %) | 5.8 ± 0.7 | 5.7 ± 0.6 | 5.7 ± 0.6 | 5.9 ± 0.7 | 0.145 |
| eGFR (mL/min/1.73 m2) | 46.6 ± 10.0 | 50.4 ± 11.8 | 41.7 ± 4.2 | 44.1 ± 7.7 | 0.001 |
| LVEF (%) | 26.6 ± 6.4 | 31.2 ± 5.5 | 23.4 ± 5.0 | 22.6 ± 4.2 | < 0.001 |
| BNP (pg/mL) | 386.0 (213.0, 650.0) | 248.0 (179.0, 371.0) | 248.0 (179.0, 371.0) | 726.0 (622.0, 793.0) | < 0.001 |
| Mean CIMT (mm) | 0.945 ± 0.145 | 0.861 ± 0.108 | 0.959 ± 0.143 | 1.036 ± 0.126 | < 0.001 |
| ACEIs (%) | 87 (76.3) | 42 (80.8) | 10 (62.5) | 35 (76.1) | 0.323 |
| ARBs (%) | 29 (25.4) | 11 (21.2) | 6 (37.5) | 12 (26.1) | 0.419 |
| β blockers (%) | 100 (87.7) | 49 (94.2) | 14 (87.5) | 37 (80.4) | 0.116 |
| CCBs (%) | 24 (21.1) | 13 (25.0) | 3 (18.8) | 8 (17.4) | 0.635 |
| Furosemide (%) | 98 (86.0) | 45 (86.5) | 14 (87.5) | 39 (84.8) | 0.952 |
| Spironolactone (%) | 67 (58.8) | 35 (67.3) | 10 (62.5) | 22 (47.8) | 0.140 |
| Statins (%) | 36 (31.6) | 16 (30.8) | 6 (37.5) | 14 (30.4) | 0.859 |
| Aspirin (%) | 47 (41.2) | 17 (32.7) | 9 (56.3) | 21 (41.2) | 0.180 |
The data are expressed as the mean ± SD or median (interquartile range), CONUT: controlling nutritional status, DCM: dilated cardiomyopathy, ICM: ischemic cardiomyopathy, NYHA: New York Heart Association, BMI: body mass index, BP: blood pressure, CRP: C-reactive protein, LDL: low-density lipoprotein, HDL: high-density lipoprotein, FPG: fasting plasma glucose, eGFR: estimated glomerular filtration rate, JDS: Japan Diabetes Society, LVEF: left ventricular ejection fraction, BNP: B-type natriuretic peptide, CIMT: carotid intima-media thickness, ACEIs: angiotensin-converting enzyme inhibitors, ARBs: angiotensin receptor blockers, CCBs: calcium channel blockers
Comparison of the Production of TNF-α by Monocytes Categorized by the CONUT Score
The production of monocyte TNF-α was significantly higher in patients with CONUT ≥ 3 than in those with CONUT 2 and CONUT 0–1 (TNF-α: 5.4 ± 1.2 vs. 3.6 ± 1.3 and 2.9 ± 1.1 pg/mL/106 PBMCs, respectively, p < 0.001). However, there were no significant differences in the monocyte TNF-α production between patients with CONUT 0–1 and those with CONUT 2 (2.9 ± 1.1 vs. 3.6 ± 1.13 pg/mL/106 PBMCs, p = 0.130).
Relationships between the CONUT Score and Multiple Variables
According to a multivariate linear regression analysis, the production of monocyte TNF-α and CRP (β coefficient = 0.277, p = 0.014), Hb (β coefficient = −0.224, p = 0.042), eGFR (β coefficient = 0.225, p = 0.049), and mCIMT (β coefficient = 0.204, p = 0.006) values were significantly and independently associated with the CONUT score (Table 3). Further adjusting for various medications, such as ACEIs, ARBs, β-blockers, aspirin, and furosemide, did not eliminate the association between the production of TNF-α and the CONUT score (β coefficient = 0.434, p < 0.001, Table 3).
Table 3. A multivariate linear regression analysis of the CONUT score.
| β | P value | |
|---|---|---|
| Age (years) | 0.104 | 0.322 |
| Gender (men) | −0.080 | 0.365 |
| Body mass index (kg/m2) | 0.008 | 0.946 |
| Systolic blood pressure (mmHg) | −0.145 | 0.124 |
| Heart rate (beats/min) | −0.077 | 0.330 |
| Hemoglobin (g/dL) | −0.224 | 0.042 |
| eGFR (mL/min/1.73 m2) | 0.225 | 0.049 |
| BNP (pg/mL) | −0.048 | 0.732 |
| C-reactive protein (mg/dL) | 0.277 | 0.014 |
| LVEF (%) | −0.068 | 0.638 |
| FPG (mg/dL) | −0.062 | 0.589 |
| HbAlc (JDS: %) | 0.061 | 0.564 |
| Monocyte TNF-α (pg/mL/106 PBMCs) | 0.434 | < 0.001 |
| CCB use | 0.121 | 0.143 |
| ACEI use | 0.000 | 0.998 |
| β-blocker use | −0.049 | 0.454 |
| Spironolactone use | −0.090 | 0.287 |
| Furosemide use | −0.060 | 0.358 |
| Aspirin use | −0.021 | 0.743 |
| Statin use | −0.076 | 0.231 |
| Mean CIMT (mm) | 0.204 | 0.006 |
| Model adjusted R2 = 0.668 | ||
β: regression coefficient, eGFR: estimated glomerular filtration rate, BNP: B-type natriuretic peptide, LVEF: left ventricular ejection fraction, FPG: fasting plasma glucose, JDS: Japan Diabetes Society, TNF: tumor necrosis factor, PBMCs: peripheral blood mononuclear cells, CCB: calcium channel blockers, ACEI: angiotensin-converting enzyme inhibitors, CIMT: carotid intima-media thickness
Fig. 1 shows the relationships between the monocyte TNF-α production, CRP, Hb, eGFR, and CONUT score. The production of monocyte TNF-α and CRP levels were significantly and positively associated with the CONUT score (TNF-α: r = 0.708, p < 0.001, Fig. 1A; CRP: r = 0.679, p < 0.001, Fig. 1B), and the Hb and eGFR values were significantly and negatively related to the CONUT score (Hb: r = −0.444, p < 0.001, Fig. 1C; eGFR: r = −0.260, p = 0.005, Fig. 1D).
Fig. 1.
Relationships between the monocyte tumor necrosis factor (TNF)-α production, C-reactive protein (CRP), and hemoglobin (Hb) and estimated glomerular filtration rate (eGFR) and the CONUT score. The production of monocyte TNF-α and CRP levels were significantly and positively associated with the CONUT score (TNF-α: r = 0.798, p < 0.001, Fig. 1A; CRP: r = 0.679, p < 0.001, Fig. 1B), and the Hb and eGFR values were significantly and negatively related to the CONUT score (Hb: r = −0.444, p < 0.001; Fig. 1C, eGFR: r = −0.260, p = 0.005, Fig. 1D).
In addition, the production of monocyte TNF-α was significantly and positively correlated with the mCIMT (r = 0.446, p < 0.001).
Comparison of the Clinical Characteristics and Production of TNF-α by Monocytes in Patients who Developed Cardiac Events and those who did not
Forty-five patients experienced cardiac events, including 16 cardiac deaths and 29 readmissions for worsening CHF, with a median follow-up period of 67.5 months (range: 12–180 months).
A comparison of the clinical characteristics of patients who developed cardiac events and those who did not is shown in Table 4. Patients with cardiac events were significantly older; had lower BMI, LVEF, Hb, eGFR, serum albumin, total cholesterol, LDL cholesterol, and triglycerides values; and demonstrated a more severe NYHA class than those without events. Furthermore, patients with cardiac events had higher heart rates and plasma CRP and BNP levels than those without events. However, there were no significant differences between the groups with respect to gender, SBP, etiology of CHF (DCM or ICM), prevalence of smoking, diabetes, or hypertension.
Table 4. Comparisons of the clinical and blood chemical variables in patients with and without cardiac events.
| All (n = 114) | Event (n = 45) | No event (n = 69) | P value | |
|---|---|---|---|---|
| Age (years) | 66.0 ± 11.3 | 70.2 ± 9.6 | 63.2 ± 11.6 | 0.001 |
| Men, n (%) | 85 (74.6) | 35 (70.0) | 71 (73.2) | 0.701 |
| DCM/ICM | 88/26 | 34/11 | 54/15 | 0.821 |
| CONUT score | 2 (0, 4) | 5 (4, 6) | 0 (0, 2) | < 0.001 |
| Total cholesterol (mg/dL) | 204 ± 27 | 187 ± 22 | 216 ± 23 | < 0.001 |
| Serum albumin (g/dL) | 3.5 ± 0.5 | 3.1 ± 0.3 | 3.8 ± 0.3 | < 0.001 |
| Lymphocyte (count/mL) | 1781 ± 640 | 1327 ± 602 | 2076 ± 469 | < 0.001 |
| NYHA class (I/II/III/IV) | 0/68/46/0 | 0/12/33/0 | 0/56/13/0 | < 0.001 |
| BMI (kg/m2) | 21.9 ± 2.2 | 20.2 ± 4.0 | 23.0 ± 1.8 | < 0.001 |
| Systolic blood pressure (mmHg) | 119 ± 8 | 117 ± 8 | 119 ± 8 | 0.127 |
| Heart rate (beats/min) | 75 ± 7 | 78 ± 7 | 74 ± 7 | 0.010 |
| Smoking (%) | 22 (19.3) | 10 (22.2) | 12 (17.4) | 0.629 |
| Hypertension (%) | 36 (31.6) | 12 (26.7) | 24 (34.8) | 0.414 |
| Diabetes (%) | 28 (24.6) | 13 (26.0) | 23 (23.7) | 0.840 |
| Dyslipidemia (%) | 53 (46.5) | 15 (33.3) | 38 (55.1) | 0.034 |
| Hemoglobin (g/dL) | 12.2 ± 1.3 | 11.7 ± 1.2 | 12.6 ± 1.3 | < 0.001 |
| CRP (mg/dL) | 0.43 (0.34, 0.63) | 0.67 (0.58, 0.71) | 0.36 (0.24, 0.43) | < 0.001 |
| Monocyte TNF-α (pg/mL/106 PBMCs) | 3.3 (2.7, 5.3) | 5.5 (4.9, 6.2) | 2.8 (2.5, 3.3) | < 0.001 |
| LDL-C (mg/dL) | 130 ± 23 | 117 ± 18 | 139 ± 21 | < 0.001 |
| HDL-C (mg/dL) | 45 ± 6 | 45 ± 5 | 46 ± 6 | 0.631 |
| Triglycerides (mg/dL) | 140 (122, 154) | 122 (102, 139) | 144 (133, 166) | < 0.001 |
| Fasting plasma glucose (mg/dL) | 106 ± 16 | 109 ± 17 | 104 ± 15 | 0.110 |
| HbAic (JDS; %) | 5.8 ± 0.7 | 5.9 ± 0.7 | 5.7 ± 0.7 | 0.114 |
| eGFR (mL/min/1.73 m2) | 46.6 ± 10.0 | 43.4 ± 7.0 | 48.8 ± 11.2 | 0.005 |
| LVEF (%) | 26.6 ± 6.4 | 22.2 ± 4.0 | 29.5 ± 6.1 | < 0.001 |
| BNP (pg/mL) | 436.5 (217.5, 727.0) | 727.0 (630.0, 794.0) | 279.0 (179.0, 404.0) | < 0.001 |
| Mean CIMT (mm) | 0.945 ± 0.145 | 1.042 ± 0.129 | 0.882 ± 0.117 | < 0.001 |
| ACEIs (%) | 87 (76.3) | 34 (75.6) | 53 (76.8) | 0.877 |
| ARBs (%) | 29 (25.4) | 12 (26.7) | 17 (24.6) | 0.829 |
| β-blockers (%) | 100 (87.7) | 36 (80.0) | 64 (92.8) | 0.077 |
| CCBs (%) | 24 (21.1) | 8 (17.8) | 16 (23.2) | 0.639 |
| Furosemide (%) | 98 (86.0) | 38 (84.4) | 60 (87.0) | 0.785 |
| Spironolactone (%) | 67 (58.8) | 21 (46.7) | 46 (66.7) | 0.051 |
| Statins (%) | 36 (31.6) | 14 (31.1) | 22 (31.9) | 0.931 |
| Aspirin (%) | 47 (41.2) | 20 (44.4) | 27 (39.1) | 0.697 |
The data are expressed as the mean ± SD or median (interquartile range), DCM: dilated cardiomyopathy, ICM: ischemic cardiomyopathy, CONUT: controlling nutritional status, BMI: body mass index, BW: body weight, CRP: C-reactive protein, LDL: low-density lipoprotein, NYHA: New York Heart Association, C: cholesterol, HDL: high density lipoprotein, eGFR: estimated glomerular filtration rate, JDS: Japan Diabetes Society, LVEF: left ventricular ejection fraction, BNP: B-type natriuretic peptide, CIMT: carotid intima-media thickness, ACEIs: angiotencin-converting enzyme inhibitors, ARBs: angiotensin receptor blockers, CCBs: calcium channel blockers
The mCIMT values were significantly higher in patients with cardiac events than in those without cardiac events (1.042 ± 0.129 vs. 0.822 ± 0.177 mm, p < 0.001). In addition, the production of monocyte TNF-α was significantly higher in patients who developed cardiac events than in those who did not (5.6 ± 1.1 vs. 3.0 ± 1.0 pg/mL/106 PBMCs, p < 0.001).
Relationships between mCIMT and Multiple Variables
According to a multivariate linear regression analysis, the production of monocyte TNF-α and CRP (β coefficient = 0.491, p = 0.001) and serum albumin (β coefficient = −0.425, p = 0.019) values were significantly and independently associated with mCIMT (Table 5). Further adjusting for various medications, such as ACEIs, ARBs, β-blockers, aspirin, and furosemide, did not eliminate the association with the production of TNF-α or mCIMT (β coefficient = 0.533, p = 0.017, Table 5).
Table 5. A multivariate linear regression analysis of the mean carotid intima-media thickness.
| β | P value | |
|---|---|---|
| Age (years) | −0.212 | 0.202 |
| Gender (men) | −0.090 | 0.497 |
| Body mass index (kg/m2) | −0.296 | 0.085 |
| Systolic blood pressure (mmHg) | −0.046 | 0.757 |
| Heart rate (beats/min) | 0.035 | 0.801 |
| Lymphocyte (count/mL) | −0.032 | 0.847 |
| Hemoglobin (g/dL) | 0.090 | 0.664 |
| Serum albumin (g/dL) | −0.425 | 0.019 |
| LDL-cholesterol (mg/dL) | −0.115 | 0.334 |
| HDL-cholesterol (mg/dL) | −0.173 | 0.136 |
| Triglycerides (mg/dL) | 0.001 | 0.990 |
| eGFR (mL/min/1.73 m2) | −0.035 | 0.853 |
| BNP (pg/mL) | 0.032 | 0.871 |
| C-reactive protein (mg/dL) | 0.491 | 0.001 |
| LVEF (%) | 0.141 | 0.505 |
| FPG (mg/dL) | −0.179 | 0.267 |
| HbAlc (JDS: %) | 0.070 | 0.656 |
| Monocyte TNF-α (pg/mL/106 PBMCs) | 0.533 | 0.017 |
| CCB use | −0.075 | 0.506 |
| ACEI use | 0.154 | 0.738 |
| β-blocker use | 0.046 | 0.630 |
| Spironolactone use | 0.144 | 0.255 |
| Furosemide use | 0.086 | 0.383 |
| Aspirin use | 0.005 | 0.958 |
| Statins use | −0.087 | 0.347 |
| Model adjusted R2 = 0.329 | ||
β: regression coefficient, LDL: low-density lipoprotein, HDL: high-density lipoprotein, eGFR: estimated glomerular filtration rate, BNP: B-type natriuretic peptide, LVEF: left ventricular ejection fraction, FPG: fasting plasma glucose, JDS: Japan Diabetes Society, TNF: tumor necrosis factor, PBMCs: peripheral blood mononuclear cells, CCB: calcium channel blockers, ACEI: angiotensin-converting enzyme inhibitors
Effects of Medications in CHF Patients with or without Cardiac Events
The prevalence of the use of β-blockers (80.0% vs. 92.8%, p = 0.077) and spironolactone (46.7% vs. 66.7%, p = 0.051) tended to be lower in patients with cardiac events than in those without cardiac events (Table 4). However, the use of ACEIs, ARBs, furosemide, CCBs, aspirin, and statins was similar between the two groups (Table 4). In addition, a multivariate Cox hazard analysis showed that the use of β-blockers and spironolactone was not associated with the incidence of cardiac events (Table 6).
Table 6. Results of the univariate and multivariate analyses for predicting cardiac events.
| Univariate |
Multivariate |
|||||
|---|---|---|---|---|---|---|
| HR | 95% C.I. | P value | HR | 95% C.I. | P value | |
| Age ≥ 70 (years) | 3.26 | 1.73–6.14 | < 0.001 | 1.69 | 0.44–6.45 | 0.442 |
| Gender (men) | 0.84 | 0.45–1.58 | 0.584 | |||
| Dilated cardiomyopathy | 0.95 | 0.48–1.87 | 0.873 | |||
| Active smoking | 1.33 | 0.65–2.69 | 0.434 | |||
| Body mass index ≤ 21 kg/m2 | 7.07 | 3.56–14.04 | < 0.001 | 1.41 | 0.42–4.75 | 0.584 |
| Systolic blood pressure ≤ 120 mmHg | 1.56 | 0.86–2.69 | 0.140 | |||
| Heart rate ≥ 78 beats/min | 1.68 | 0.93–3.03 | 0.086 | |||
| CONUT score ≥ 3 | 44.55 | 10.76–184.37 | < 0.001 | 11.97 | 2.21–64.67 | 0.004 |
| Monocyte TNF-α ≥ 4.1 pg/mL/106 PBMCs | 29.46 | 9.10–95.32 | < 0.001 | 14.10 | 2.55–77.92 | 0.002 |
| LVEF ≤ 26% | 4.78 | 2.40–9.49 | < 0.001 | 2.31 | 0.57–9.34 | 0.240 |
| BNP ≥ 600 pg/mL | 8.66 | 4.02–18.64 | < 0.001 | 1.53 | 0.38–6.26 | 0.553 |
| FPG ≥ 105 mg/dL | 1.50 | 0.84–2.70 | 0.176 | |||
| C-reactive protein ≥ 0.47 mg/dL | 15.27 | 5.99–38.50 | < 0.001 | 2.12 | 0.51–8.88 | 0.304 |
| eGFR ≤ 45 mL/min/1.73 m2 | 2.62 | 1.40–4.91 | 0.003 | 1.04 | 0.35–3.13 | 0.940 |
| Hemoglobin ≤ 12.0 g/dL | 3.50 | 1.86–6.60 | < 0.001 | 1.49 | 0.30–7.45 | 0.626 |
| Calcium channel blocker use | 0.62 | 0.29–1.34 | 0.222 | |||
| ACEI use | 0.79 | 0.40–1.56 | 0.492 | |||
| β-blocker use | 0.34 | 0.16–0.71 | 0.004 | 0.82 | 0.31–2.14 | 0.677 |
| Furosemide use | 0.80 | 0.36–1.80 | 0.591 | |||
| Spironolactone use | 0.52 | 0.29–0.93 | 0.029 | 0.70 | 0.23–2.15 | 0.529 |
| Aspirin use | 1.11 | 0.61–1.99 | 0.739 | |||
| Statin use | 0.81 | 0.43–1.52 | 0.507 | |||
| Mean CMIT ≥ 1.0 mm | 4.69 | 2.58–8.54 | < 0.001 | 1.45 | 0.69–3.04 | 0.328 |
HR: hazard ratio, C.I.: confidence interval, CONUT: controlling nutritional status, TNF: tumor necrosis factor, PBMCs: peripheral blood mononuclear cells, LVEF: left ventricular ejection fraction, BNP:B-type natriuretic peptide, FPG: fasting plasma glucose, eGFR: estimated glomerular filtration ratio, ACEI: angiotensin-converting inhibitors, CIMT: carotid intima-media thickness
ROC Curve Analysis
The ROC curve analysis revealed that the optimal cut-off values for predicting cardiac events with respect to the BMI, CONUT score, serum albumin, production of monocyte TNF-α, age, heart rate, LVEF, Hb, eGFR, mCIMT, BNP, and CRP were 21.0 kg/m2, 3, 3.3 g/dL, 4.1 pg/mL/106 PBMCs, 70 years, 78 beats/min, 26%, 12.0 g/dL, 45.0 ml/min/1.73 m2, 1.0 mm, 600 pg/mL, and 0.47 mg/dL, respectively.
The sensitivities and specificities for the BMI, CONUT score, serum albumin, production of monocyte TNF-α, subject age, heart rate, LVEF, Hb, eGFR, mCIMT, BNP, and CRP levels for predicting cardiac events were assessed across a range of cut-off values. The optimal cut-off value for each parameter was obtained at a sensitivity of 87.0%, 96.4%, 79.8%, 94.6%, 58.9%, 57.1%, 87.6%, 78.4%, 55.7%, 75.6%, 85.7%, and 82.1%, respectively, and a specificity of 75.6%, 95.9%, 90.8%, 89.7%, 64.9%, 63.9%, 71.4%, 64.3%, 66.1%, 78.3%, 85.6%, and 88.7%, respectively.
Relationships between the Production of Monocyte TNF-α and the CONUT Score and Cardiac Events
Fig. 2 shows the distribution of the CONUT scores among patients with (black columns) or without (open columns) cardiac events. Patients with cardiac events had higher CONUT scores than those without cardiac events. During a median follow-up period of 67.5 months (range: 12–180 months), cardiac events, including 16 cardiac deaths and 27 readmissions due to worsening CHF, occurred in 43 of the 46 patients (93.5%) with a CONUT score of ≥ 3, and such events (two readmissions for worsening CHF without death) occurred in only two of the 16 patients (12.5%) with a CONUT score of 2, whereas none of the patients with a CONUT score of 0–1 developed cardiac events (Fig. 3A). The incidence of cardiac events in patients with a CONUT score of ≥ 3 was significantly higher than that observed in those with a CONUT score of 2 (log-rank 14.06, p < 0.001). In addition, the incidence of events in patients with a CONUT score of 2 was significantly higher than that observed in those with a CONUT score of 0–1 (logrank 6.43, p = 0.011, Fig. 3A).
Fig. 2.
Distribution of the CONUT scores in CHF patients with cardiac events (black columns) and those without cardiac events (open columns).
Fig. 3.
Kaplan–Meier event-free curves and log-rank tests indicated a CONUT score of 0–1, CONUT score of 2, and CONUT score of ≥ 3 to be significantly related to an increased rate of cardiac event rates with an increasing CONUT score during the follow-up period (CONUT score 0–1 vs. CONUT score 2, log-rank 6.43, p = 0.011 and CONUT score 2 vs. CONUT score ≥ 3, log-rank 14.06, p < 0.001, Fig. 3A). In comparison with patients with TNF-α production of < 4.1 pg/mL/106 PBMCs, patients with a monocyte TNF-α production of ≥ 4.1 g/mL/106 PBMCs were significantly more likely to exhibit an increased rate of cardiac events during the follow-up period (log-rank 76.97, p < 0.001, Fig. 3B). Patients with a low serum albumin level (≤ 3.3 g/dL) exhibited a significantly lower event-free survival than those with a high serum albumin level (> 3.3 g/dL, log-rank 61.32, p < 0.001, Fig. 3C). PBMCs: peripheral blood mononuclear cells.
Furthermore, cardiac events occurred in 42 of the 45 patients (14 cardiac deaths and 28 readmissions for worsening CHF; 93.3%) with elevated TNF-α production (TNF-α ≥ 4.1 pg/mL/106 PBMCs), whereas cardiac events were noted in only three of the 69 patients (two cardiac deaths and one readmission due to worsening CHF; 4.3%) without elevated TNF-α production (TNF-α < 4.1 pg/mL/106 PBMCs, log-rank 76.97, p < 0.001, Fig. 3B).
In addition, patients with a lower serum level of albumin (≤ 3.3 g/dL) experienced 32 cardiac events (94.1%, 13 cardiac deaths and 19 readmissions for worsening CHF) in comparison with 13 events (17.4%, three cardiac deaths and 10 readmissions due to worsening CHF) observed in patients with a serum albumin level of > 3.3 g/dL (log-rank: 61.13, p < 0.001, Fig. 3C).
Cox Proportional Hazard Analysis of Cardiac Events
The results of both univariate and multivariate Cox hazard analyses are shown in Table 6. A multivariate analysis showed that a CONUT score of ≥ 3 (HR, 11.97; 95% CI, 2.21 -64.67, p = 0.004) and monocyte TNF-α level of ≥ 4.1 pg/mL/106 PBMCs (HR, 14.10; 95% CI, 2.55–77.92; p = 0.002) were each significantly and independently associated with cardiac events.
Discussion
This study provides evidence that malnutrition as assessed using the CONUT score is common and predicts a poor outcome in patients with CHF. Furthermore, patients with cardiac events exhibit a significantly higher production of monocyte TNF-α, mCIMT, and plasma levels of CRP and lower serum albumin, Hb, and eGFR values than those without cardiac events, and there are significant positive correlations between the production of monocyte TNF-α, mCIMT, and plasma levels of CRP and the CONUT score. In addition, the renal function, estimated by eGFR, and anemia are also related to the CONUT score. Taken together, the CONUT score is significantly associated with inflammation, carotid atherosclerosis indicated by mCIMT, the renal function, and anemia in patients with CHF.
In this study, a high CONUT score (CONUT score ≥ 3) as well as the upregulation of monocyte TNF-α production predicted a poor outcome in patients with CHF. Therefore, we herein provide the first evidence that malnutrition assessed using the CONUT score is significantly associated with the upregulation of monocyte-derived TNF-α production, carotid atherosclerosis as indicated by mCIMT, the renal function, and anemia and predicts a poor outcome in patients with CHF.
Mechanisms Involved in the Upregulation of Monocyte Proinflammatory Cytokine Production
The mechanisms by which proinflammatory cytokines, including TNF-α, produced by monocytes are increased in CHF patients remain unclear. However, the production of monocyte TNF-α may be enhanced by various stimuli, such as angiotensin II, CRP, and endotoxin (LPS: lipopolysaccharide) and proinflammatory cytokines, including TNF-α itself1–3, 19–21)
A significant decrease in the intestinal blood flow has been observed at low levels of exercise in CHF patients, resulting in intestinal ischemia22). Inadequate mucosal perfusion increases intestinal mucosal permeability; thus, LPS may be able to enter the circulation through the gut wall if the barrier function is impaired, as observed in various diseases, including sepsis and CHF. In CHF patients, LPS may activate monocytes and macrophages in the circulation, thereby upregulating the production of proinflammatory cytokines, including TNF-α21).
In our previous study, the endotoxin (LPS) levels were measured in 20 normal subjects and 40 patients with CHF23). The endotoxin levels in patients with CHF (median: 30.5 pg/mL, range: 11.0–95.2 pg/mL) were significantly higher than those observed in normal subjects (median: 11.8 pg/mL; range 10.2–18.9 pg/mL, p < 0.001). In addition, the production of monocyte TNF-α was significantly and positively associated with the plasma levels of endotoxin (r = 0.803, p < 0.001)23). These data suggest that the upregulation of monocyte TNF-α production may be, in part, because of increases in LPS from the plasma.
Furthermore, we have previously reported that the plasma CRP levels tend to be elevated and associated with the production of monocyte TNF-α in CHF patients17). Interestingly, CRP upregulates the production of monocyte TNF-α24). Therefore, increased plasma CRP levels may lead to increased production of monocyte TNF-α in CHF patients. Taken together, these findings indicate that LPS and CRP in the plasma as well as upregulated monocyte-derived proinflammatory cytokines themselves and combinations of these factors may upregulate the production of monocyte cytokines in CHF patients.
Cardio-Renal Anemia Syndrome and Malnutrition
Chronic kidney disease (CKD) and anemia often occur during the course of CHF and induce activation of the sympathetic nervous system and renin–angiotensin–aldosterone system as well as oxidative stress and inflammation25, 26). Silveberg et al.27) referred to the association among CHF, renal dysfunction, and anemia as cardio-renal anemia syndrome, where CHF may cause progressive renal dysfunction, both of which may lead to anemia, which in turn can worsen CHF and renal dysfunction. This study showed that renal dysfunction and anemia are associated with poor outcomes in CHF patients.
Combined disorders of the heart and kidney are likely to develop in the presence of some degree of cachexia and are associated with organ crosstalk via TNF-α28). Under these circumstances, a vicious cycle may arise in which cachexia and malnutrition associated with either CHF or CKD may contribute to further damage28). This study showed that anemia and renal dysfunction estimated by eGFR are significantly and independently related to the CONUT score. Therefore, cardio-renal anemia syndrome may be associated with malnutrition in CHF patients.
Cardiac Cachexia (Malnutrition) is Significantly Related to Inflammation
Cachexia is associated with progressive muscle wasting and the loss of fat mass and can be observed in several different chronic disorders, including CHF29). However, there is no common definition of “cachexia” so far. A cut-off level of at least 10% loss of lean tissue has been used to define cardiac cachexia30). As this definition requires specialized tools to assess body composition, such as dual energy X-ray absorptiometry, this may not be a practical definition. As a better alternative, Springer and Akashi et al.29) defined cardiac cachexia as the occurrence of a non-edematous weight loss of > 6% over a period of 6 months, regardless of the body mass index.
Cardiac cachexia has been identified to be an independent risk factor for mortality in patients with CHF1, 2, 31) and is associated with both hypoalbuminemia (< 3.5 g/dL) and muscle atrophy32). The complex mechanisms of the progression from heart failure to cardiac cachexia are not fully understood and multiple pathways may be involved. The activation of inflammatory pathways plays an important role in all forms of cachexia. TNF-α is one of the key cytokines in the development of catabolism, together with interleukin (IL)-1 and IL-629).
The TNF-α values have been reported to be significantly increased in patients with CHF cachexia9), indicating that cardiac cachexia is related to inflammation. The onset of CHF cachexia associated with chronic inflammation results from bacterial or endotoxin translocation due to bowel wall edema following severe heart failure20–22).
The mechanism underlying the development of protein–energy malnutrition (PEM) in CHF patients may involve proinflammatory cytokine activation associated with increased endotoxin absorption and/or reduced clearance1). PEM and inflammation are also strongly associated with one another and may each independently contribute to hypoalbuminemia and cachexia in patients with CHF1). These data suggest that malnutrition associated with inflammation may contribute to the pathogenesis and development of CHF.
Recently, various assessment tools and risk indices have been developed in accordance with the clinical setting and timing. The CONUT score was developed by Ignacio de Ulíbarri et al.5) as a screening tool for identifying undernutrition in hospital populations. The following three parameters are used to calculate the score: the serum albumin level (g/L), total cholesterol level (mg/dL), and total lymphocyte count (count/ml). Therefore, the CONUT score reflects the level of protein reserves, caloric depletion, and immune defenses, respectively, in a given patient5).
In addition, Nochioka et al.33) showed that in current stage B patients (classified as asymptomatic subjects with structural and/or functional heart disease according to the ESC/ACC/AHA guidelines34), a poor nutritional status, as indicated by the CONUT score, is significantly associated with an increased incidence of death in the overall population and heart failure hospitalization among the elderly. However, the pathophysiological mechanisms underlying these findings remain unclear.
Malnutrition, Inflammation, and Atherosclerosis Syndrome
Traditional risk factors (hypertension, dyslipidemia, and obesity) for atherosclerotic cardiovascular disease (ACD) have been shown to be associated with the incidence of heart failure35).
CIMT is generally considered to be an early indicator of subclinical atherosclerosis and may reflect the generalized ACD process of an individual that ultimately results in heart failure. Engström et al.35) showed that in the general population without a history of ACD, high CIMT and CRP values are both independent risk factors for the incidence of heart failure requiring hospitalization, indicating that systemic inflammation and carotid atherosclerosis play an important role in the pathogenesis of CHF.
This study showed that carotid atherosclerosis as indicated by mCIMT is significantly associated with the CONUT score. A multivariate linear regression analysis showed that the production of monocyte TNF-α and plasma levels of CRP were significantly and positively associated with mCIMT, and the serum levels of albumin were significantly and negatively related to mCIMT. In addition, as mentioned above, malnutrition is significantly associated with inflammation and renal disease. Taken together, malnutrition as indicated by the CONUT score is significantly related to inflammation and carotid atherosclerosis (malnutrition, inflammation, and atherosclerosis syndrome).
Lower serum albumin levels are associated with an increased risk of cardiovascular mortality, coronary artery disease, and stroke, suggesting protective effects of albumin against atherosclerosis36). Ishizawa et al.37) showed that when compared with the lowest quartile, the highest quartile of serum albumin was associated with a reduced prevalence of carotid plaque and CIMT in Japanese individuals. In addition, Lapenna et al.38) showed that the serum albumin levels were significantly and inversely correlated with those of atherosclerotic plaque thiobarbituric acid reactive substances and advanced oxidation protein products as indicated by oxidative damage in human atherosclerotic plaques, indicating that serum albumin possesses antioxidant properties.
This study showed that the serum albumin levels were significantly and negatively related to mCIMT. Inflammation and oxidative stress play an important role in the pathogenesis and exacerbation of CHF. Taken together, inflammation and low serum albumin may play a significant role in the progression of carotid atherosclerosis in patients with CHF.
In addition, this study showed that cardiac events, including cardiac deaths and re-hospitalization due to worsening CHF, occurred due to severe pump failure but not the development of acute coronary syndrome. We enrolled patients with ICM who had received CABG or PCI before enrollment and had no evidence of myocardial ischemia using a treadmill exercise test or exercise myocardial perfusion scintigraphy at entry. In addition, they did not experience chest pain during the follow-up period, and the follow-up treadmill exercise test or exercise perfusion myocardial scintigraphy showed no evidence of myocardial ischemia during the follow-up period. Therefore, patients with ICM may not develop acute myocardial infarction or unstable angina during the follow-up period.
Nutrition and Neurohormones in Heart Failure
Numerous hormone systems, including those involving ghrelin and leptin, may contribute to the wasting process by altering appetite and energy expenditure. The derangement of these hormone systems, potentially triggered by the effects of proinflammatory cytokines, such as TNF-α, may be responsible for the development of satiety without an adequate food intake39, 40).
As mentioned above, malabsorption from the gut as a result of bowel edema and decreased bowel perfusion due to heart failure may also play a pivotal role in the progression of the wasting process. Further studies are needed to clarify the precise mechanisms underlying the wasting pathways.
This study showed that malnutrition is very common and significantly associated with anemia, the renal function, and atherosclerosis as indicated by mCIMT and inflammation in patients with CHF. In addition, malnutrition as indicated by the CONUT score predicts a poor outcome in patients with CHF.
Study Limitations
There are several limitations associated with this study. First, the sample size was small. Second, the use of medications was not assigned in a randomized manner. Third, patients with CHF received β-blockers, ACEIs, furosemide, or spironolactone for at least 3 years prior to study enrollment, and those in the acute decompensated stage of CHF were excluded. Therefore, all the eligible patients in this study had been stabilized with a standard treatment prior to enrollment, and the effects of the proinflammatory cytokine levels and nutritional status may thus differ in unstable CHF patients. Fourth, we did not enroll consecutive CHF patients because we had to exclude patients in the acute decompensated stage of CHF; therefore, we only selected stable CHF patients at out-patient clinics. Fifth, the definition of cardiac cachexia remains controversial; however, we did not have any data on cardiac cachexia in this study.
Conclusion
In conclusion, malnutrition as assessed using the CONUT score is a common phenomenon and predicted a poor outcome in patients with CHF. Furthermore, patients with cardiac events exhibited a significantly higher production of monocyte TNF-α, plasma levels of CRP, and mCIMT and lower Hb, serum albumin, and eGFR values than those without cardiac events, and there were significant positive correlations between the production of monocyte TNF-α, plasma levels of CRP, and mCIMT and the CONUT score. In addition, the renal function, estimated by eGFR, and anemia were also related to the CONUT score. Taken together, the CONUT score was significantly associated with inflammation and atherosclerosis as indicated by CIMT, the renal function, and anemia in patients with CHF.
Further studies are needed to elucidate the precise mechanisms involved in this pathophysiological pathway and develop new therapeutic strategies for preventing the deterioration of CHF.
Acknowledgements
None.
Disclosure
None.
References
- 1). Kalantar-Zadeh K, Block G, Horwich T, Fonarow GC., El-Menyar AA. Reverse epidemiology of conventional cardiovascular risk factors in patients with chronic heart failure. J Am Coll Cardiol, 2004; 43: 1439-1444 [DOI] [PubMed] [Google Scholar]
- 2). Sargento L, Longo S, Lousada N, dos Reis RP. The importance of assessing nutritional status in elderly patients with heart failure. Curr Heart Dail Rep, 2014; 11: 220-226 [DOI] [PubMed] [Google Scholar]
- 3). Otaki M. Surgical treatment of patients with cardiac cachexia. An analysis of factors affecting operative mortality. Chest, 1993; 105: 1347-1351 [DOI] [PubMed] [Google Scholar]
- 4). Tevik K, Thürmer H, Husby MI, de Soysa AK, Helvik AS. Nutritional risk screening in hospitalized patients with heart failure. Clin Nutr, 2014; 34: 257-264 [DOI] [PubMed] [Google Scholar]
- 5). Ignacio de Ulíbarri J, González-Madroño A, de Villar NG, González P, González B, Mancha A, Rodríguez F, Fernández G. CONUT: a tool for controlling nutritional status. First validation in a hospital population. Nutr Hosp, 2005; 20: 38-45 [PubMed] [Google Scholar]
- 6). El-Menyar AA. Cytokines and myocardial dysfunction: State of the art. J Cardiac Fail, 2008; 14: 61-74 [DOI] [PubMed] [Google Scholar]
- 7). Mann DL. Inflammatory mediators and the failing heart. Past, present, and the foreseeable future. Circ Res, 2002; 91: 988-998 [DOI] [PubMed] [Google Scholar]
- 8). Hirasawa Y, Nakagomi A, Kobayashi Y, Katoh T, Mizuno K. Short-term amiodarone treatment attenuates the production of monocyte cytokines and chemokines by C-reactive protein and improves cardiac function in patients with idiopathic dilated cardiomyopathy and ventricular tachycardia. Circ J, 2009; 73: 639-646 [DOI] [PubMed] [Google Scholar]
- 9). Levine B, Kalman J, Mayer L, Fillit HM, Packer M. Elevated circulating levels of tumor necrosis factor in severe chronic heart failure. N Engl J Med, 1990; 323: 236-241 [DOI] [PubMed] [Google Scholar]
- 10). Hasper D, Hummel M, Kleber FX, Reindl I, Volk HD. Systemic inflammation in patients with heart failure. Eur Heart J, 1998; 19: 761-765 [DOI] [PubMed] [Google Scholar]
- 11). Lorenz MZ, Markus HS, Bots ML, Rosvall M, Sitzer M. Prediction of clinical cardiovascular events with carotid intima-media thickness: a systemic review and meta-analysis. Circulation, 2007; 115: 459-467 [DOI] [PubMed] [Google Scholar]
- 12). Adams MR, Nakagomi A, Keech A, Robinson J, McCredie R, Bailey BP, Freedman SB, Celermajer DS. Carotid intima-media thickness is only weakly correlated with the extent and severity of coronary artery disease. Circulation, 1995; 92: 2127-2134 [DOI] [PubMed] [Google Scholar]
- 13). Cao JJ, Arnold AM, Manolio TA, Polak JF, Psaty BM, Hirsch CH, Kuller LH, Cushman M. Association of carotid artery intima-media thickness, plaques, and C-reactive protein with future cardiovascular disease and all-cause mortality. The cardiovascular health study. Circulation, 2007; 116: 32-38 [DOI] [PubMed] [Google Scholar]
- 14). Suzuki J, Kurosu T, Kon T, Tomaru T. Impact of cardiovascular risk factors on progression of arteriosclerosis in younger patients: evaluation by carotid duplex ultrasonography and cardio-ankle vascular index(CAVI). J Atheroscler Thromb, 2014; 21: 554-562 [PubMed] [Google Scholar]
- 15). Su TC, Liao CC, Chien KL, Hsu SH, Sung FC. An over-weight or obese status in childhood predicts subclinical atherosclerosis and prehypertension/hypertension in young adults. J Atheroscler Thromb, 2014; 21: 1170-1182 [DOI] [PubMed] [Google Scholar]
- 16). Nakagomi A, Freedman SB, Gezcy CL. Interferon-γ and lipopolysaccharide potentiate monocyte tissue factor induction by C-reactive protein. Relationship with age, sex, and hormone replacement treatment. Circulation, 2000; 101: 1785-1791 [DOI] [PubMed] [Google Scholar]
- 17). Nakagomi A, Seino Y, Endoh Y, kusama Y, Atarashi H, Mizuno K. Upregulation of monocyte proinflammatory production by C-reactive protein is significantly related to ongoing myocardial damage and future cardiac events in patients with chronic heart failure. J Cardiac Fail, 2010; 16: 562-571 [DOI] [PubMed] [Google Scholar]
- 18). Imai E, Horio M, Watanabe T, Iseki K, Yamagata K, Hara S, Ura N, Kiyohara Y, Moriyama T, Ando Y, Fujimoto S, Konta T, Yokoyama H, Makino H, Hishida A, Matsuo S. Prevalence of chronic kidney disease in the Japanese general population. Clin Exp Nephrol, 2009; 13: 621-630 [DOI] [PubMed] [Google Scholar]
- 19). Niebauer J, Volk HD, Kemp M, Dominguez M, Schumann RR, Rauchhaus M, Poole-Wilson PA, Coats AJ, Anker SD. Endotoxin and immune activation in chronic heart failure: a prospective cohort study. Lancet, 1999; 353: 1838-1842 [DOI] [PubMed] [Google Scholar]
- 20). Stoll LL., Denning GM, Weintraub NL. Potential role of endotoxin as a proinflammatory mediator of atherosclerosis. Arterioscler Thromb Vasc Biol, 2004; 24: 2227-2236 [DOI] [PubMed] [Google Scholar]
- 21). Rauchhaus M, Coats AJS, Anker SD. The endotoxinlipoprotein hypothesis. Lancet, 2000; 356 930-933 [DOI] [PubMed] [Google Scholar]
- 22). Sandek A, Bauditz J, Swidsinski A, Buhner S, Weber-Eibel J, von Haehling S, Schroedl W, Karhausen T, Doehner W, Rauchhaus M, Poole-Wilson P, Volk HD, Lochs H, Anker SD. Altered intestinal function in patients with chronic heart failure. J Am Coll Cardiol, 2007; 50: 1561-1569 [DOI] [PubMed] [Google Scholar]
- 23). Nakagomi A, Seino Y, Noma S, Kohashi K, Kosugi M, Kato K, Kusama Y, Atarashi H, Shimizu W. Relationships between the serum cholesterol levels, production of monocyte proinflammatory cytokines and long-term prognosis in patients with chronic heart failure. Intern Med, 2015; 53: 2415-2424 [DOI] [PubMed] [Google Scholar]
- 24). Ballou SP, Lozanski G. Induction of inflammatory cytokine release from cultured human monocytes by C-reactive protein. Cytokine, 1992; 4: 361-368 [DOI] [PubMed] [Google Scholar]
- 25). Bongartz LG, Cramer MJ, Doevendans PA, Joles JA, Braam B. The severe cardiorenal syndrome: ‘Guyton revisited’. Eur Heart J, 2005; 26: 11-17 [DOI] [PubMed] [Google Scholar]
- 26). Scrutinio D, Passantino A, Santoro D, Catanzaro R. The cardiorenal anaemia syndrome in systolic heart failure: prevalence, clinical correlates, and long-term survival. Eur J Heart Fail, 2011; 13: 61-67 [DOI] [PubMed] [Google Scholar]
- 27). Silverberg DS, Wexler D, Iaina A, Steinbruch S, Wollman Y, Schwartz D. Anemia, chronic renal disease and congestive heart failure--the cardio renal anemia syndrome: the need for cooperation between cardiologists and nephrologists. Int Urol Nephrol, 2006; 38: 295-310 [DOI] [PubMed] [Google Scholar]
- 28). Ronco C, Cicoira M, McCullough PA. Cardiorenal syndrome type 1: pathophysiological crosstalk leading to combined heart and kidney dysfunction in the setting of acutely decompensated heart failure. J Am Coll Cardiol, 2012; 60: 1031-1042 [DOI] [PubMed] [Google Scholar]
- 29). Springer J, Filippatos G, Akashi YJ, Anker SD. Prognosis and therapy approaches of cardiac cachexia. Curr Opin Cardiol, 2006; 21: 229-233 [DOI] [PubMed] [Google Scholar]
- 30). Anker SD, Coats AJ. Cardiac cachexia: a syndrome with impaired survival and immune and neuroendocrine activation. Chest, 1999; 115: 836-847 [DOI] [PubMed] [Google Scholar]
- 31). Ajayi AA, Adigun AQ, Ojofeitimi EO, Yusuph H, Ajayi OE. Anthropometric evaluation of cachexia in chronic congestive heart failure: the role of tricuspid regurgitation. Int J Cardiol, 1999; 71: 79-84 [DOI] [PubMed] [Google Scholar]
- 32). Pasini E, Aquilani R, Gheorghiade M, Dioguardi FS. Malnutrition, muscle wasting and cachexia in chronic heart failure: the nutritional approach. Ital Heart J, 2003; 4: 232-235 [PubMed] [Google Scholar]
- 33). Nochioka K, Sakata Y, Takahashi J, Miyata S, Miura M, Takada T, Fukumoto Y, Shiba N, Shimokawa H, CHART-2 Investigators Prognostic impact of nutritional status in asymptomatic patients with cardiac diseases. A report from the CHART-2 Study. Circ J, 2013; 77: 2318-2326 [DOI] [PubMed] [Google Scholar]
- 34). McMurray JJ, Adamopoulos S, Anker SD, Auricchio A, Böhm M, Dickstein K, Falk V, Filippatos G, Fonseca C, Gomez-Sanchez MA, Jaarsma T, Køber L, Lip GY, Maggioni AP, Parkhomenko A, Pieske BM, Popescu BA, Rønnevik PK, Rutten FH, Schwitter J, Seferovic P, Stepinska J, Trindade PT, Voors AA, Zannad F, Zeiher A, ESC Committee for Practice Guidelines ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the ESC. Eur Heart J, 2012; 33: 1787-1847 [DOI] [PubMed] [Google Scholar]
- 35). Engström G, Melander O, Hedblad B. Carotid intimamedia thickness, systemic inflammation, and incidence of heart failure hospitalizations. Arterioscler Thromb Vasc Biol, 2009; 29: 1691-1695 [DOI] [PubMed] [Google Scholar]
- 36). Nelson JJ, Liao D, Sharrett AR, Folsom AR, Chambless LE, Shahar E, Szklo M, Eckfeldt J, Heiss G. Serum albumin level as a predictor of incident coronary heart disease: the Atherosclerosis Risk in Communities (ARIC) study. Am J Epidemiol, 2000; 15: 468-477 [DOI] [PubMed] [Google Scholar]
- 37). Ishizaka N, Ishizaka Y, Nagai R, Toda E, Hashimoto H, Yamakado M. Association between serum albumin, carotid atherosclerosis, and metabolic syndrome in Japanese individuals. Atherosclerosis, 2007; 193: 373-379 [DOI] [PubMed] [Google Scholar]
- 38). Lapenna D, Ciofani G, Ucchino S, Pierdomenico SD, Cuccurullo C, Giamberardino MA, Cuccurullo F. Serum albumin and biomolecular oxidative damage of human atherosclerotic plaques. Clin Biochem, 2010; 43: 1458-1460 [DOI] [PubMed] [Google Scholar]
- 39). von Haehling S, Doehner W, Anker SD. Nutrition, metabolism, and the complex pathophysiology of cachexia in chronic heart failure. Cardiovasc Res, 2007; 73: 298-309 [DOI] [PubMed] [Google Scholar]
- 40). Fudim M, Wagman G, Altschul R, Yucel E, Bloom M, Vittorio TJ. Pathophysiology and treatment options for cardiac anorexia. Curr Heart Fail Rep, 2011; 8: 147-153 [DOI] [PubMed] [Google Scholar]