Skip Navigation


European Heart Journal Advance Access originally published online on April 8, 2005
European Heart Journal 2006 27(6):691-699; doi:10.1093/eurheartj/ehi195
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
27/6/691    most recent
ehi195v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Related articles in EHJ
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (10)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Christ, M.
Right arrow Articles by Maisch, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Christ, M.
Right arrow Articles by Maisch, B.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The European Society of Cardiology 2005. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org

Prognostic significance of serum cholesterol levels in patients with idiopathic dilated cardiomyopathy

Michael Christ*, Theresia Klima, Wolfram Grimm, Hans-Helge Mueller and Bernhard Maisch

Klinik für Innere Medizin, Kardiologie und Intensivmedizin, Philipps University Marburg, Baldingerstraße, D-35033 Marburg, Germany

Received 12 August 2004; revised 28 January 2005; accepted 3 February 2005; online publish-ahead-of-print 8 April 2005.

* Corresponding author. Tel: +49 6421 286 6462; fax: +49 6421 953825. E-mail address: Christ_michael{at}yahoo.de

See page 641 for the editorial comment on this article (doi:10.1093/eurheartj/ehi812)


   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Discussion
 Acknowledgement
 References
 
Aims Previous studies indicate that low cholesterol levels are associated with adverse prognosis in heart failure patients, because elevated lipoprotein levels may negate bacterial endotoxin load induced by gastrointestinal congestion.

Methods and results We examined the prognostic significance of lipid levels in a cohort of 422 patients with idiopathic dilated cardiomyopathy (iDCM) [50±12 years, 342 males, 80 females, left ventricular ejection fraction (LV-EF): 31.6±10.6%]. During 42 months of follow-up, 86 patients (20.3%) died or received a heart transplant. In univariate Cox regression analysis, reduced LV-EF, high New York Heart Association (NYHA) class, and increased LV end-diastolic diameter (LVEDD) were strong risk factors associated with that endpoint, whereas decreased total cholesterol, HDL-cholesterol, and apoprotein I levels were identified as weak risk predictors. After step-wise multivariable analysis, only LVEDD, NYHA class, and LV-EF emerged as parameters independently contributing to the model predicting risk for death or heart transplantation (P<0.05). Cholesterol levels were positively associated with LV-EF and negatively associated with LVEDD (P<0.05). Circulating sCD14 levels, a marker of endotoxin exposure, were related to cholesterol levels (P<0.05) and LV-EF (P<0.05).

Conclusion Decreased cholesterol levels do not independently predict adverse prognosis in patients with iDCM. Our findings indicate that low cholesterol levels are dependent on the severity of cardiac disease.

Key Words: Cholesterol • Chronic heart failure • Dilated cardiomyopathy • Risk stratification


   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Discussion
 Acknowledgement
 References
 
Experimental and clinical evidence indicates that hypercholesterolaemia and inflammation are essentially involved in the pathogenesis of coronary artery disease (CAD).1 Cholesterol synthase inhibitors (statins) reduce cholesterol levels, and subsequently attenuate progression of CAD leading to reduced morbidity and mortality in CAD patients.2–5 The use of statins reportedly prevent the development of new-onset heart failure.6 However, adverse effects of hypocholesterolaemia due to statin therapy have been suggested in patients with severe heart failure.7,8 Observational studies report that low cholesterol and triglyceride levels indicate adverse prognosis in patients with chronic heart failure (CHF).8–11

Outcome in CHF patients is most notably dependent on the severity of cardiac disease. A reduced left ventricular ejection fraction (LV-EF), an increased LV end-diastolic diameter (LVEDD), and a higher New York Heart Association (NYHA) functional class have emerged as most important risk predictors associated with adverse prognosis in heart failure patients irrespective of its aetiology.12–15 Progressive systolic pump failure and incremental LV wall strain lead to an activation of neurohormones including circulating and local catecholamines, natriuretic peptides, pro-inflammatory cytokines, and the renin–angiotensin–aldosterone system. Elevation of neurohormones and cytokines have been linked to detrimental cardiac remodelling, progressive pump failure, and sudden cardiac death suggesting a vicious circle.16,17 Cytokine activation in CHF patients at least partially depends on the exposure of bacterial endotoxins due to intestinal congestion.7,18,19 Cholesterol- and triglyceride-rich lipoproteins may have the capacity to bind and detoxify bacterial endotoxins suggesting that low cholesterol levels may be deleterious due to the inability of lipoproteins to oppose endotoxin exposure.7 Indeed, overall outcome has been improved experimentally by cholesterol containing lipoproteins after endotoxin exposure.20–24

Although this ‘endotoxin–lipoprotein hypothesis’ appears conclusive, published observational studies have been small and enrolled highly selected patients.18,25,26 Moreover, CAD patients were included in those surveys and CAD by itself is an inflammatory disease.1 Therefore, we prospectively examined the independent prognostic role of lipid levels and established markers of disease severity12–15 in CHF patients with idiopathic dilated cardiomyopathy (iDCM) as a model of ‘pure’ heart failure. Our hypothesis was that lipid levels should be a predictor of death or heart transplantation in CHF patients with iDCM using a multivariable Cox regression model. Finally, we explored whether cholesterol levels are related to circulating sCD14 levels in vivo, a marker of endotoxin exposure.18,27,28


   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Discussion
 Acknowledgement
 References
 
Patients and study design
The study was approved by the institutional review board of the Marburg University, Germany, and complies with the Declaration of Helsinki. Consecutive heart failure patients with iDCM (LV-EF≤45%), who have been referred to the Department of Internal Medicine, Cardiology, for further diagnostics, were prospectively enrolled in the Marburg Cardiomyopathy database after patients had given written, informed consent. Enrolment started in December 1992 after complete non-invasive and invasive cardiac evaluation had been conducted and lasted until December 2001. We analysed data from 422 iDCM patients of this database (Figure 1A) with the date of lipid measurement used as the time of entry into this study. A histogram of patient enrolment per year is given in Figure 1B.


Figure 1951
View larger version (17K):
[in this window]
[in a new window]
  		Figure 1 Study details: consecutive patients with an ejection fraction ≤45%, which qualified for study entry (see inclusion and exclusion criteria), were screened and 527 agreed to participate in the study. Lipid levels were available in 425 patients and follow-up was complete in 422 patients (A). A histogram displaying patients enrolled in the study per year using the date of lipid measurement as the time of study entry is shown in (B).

 
All patients were stable with regard to symptoms and drug treatment for at least 1 month before enrolment. Echocardiographic examinations were performed at the time of enrolment using a Sonotron Vingmed CFM 799 machine (Sonotron, Oslo, Norway). LVEDD and ejection fraction were measured according to current recommendations.29 In addition, complete left (and in part right) heart cardiac catheterization including coronary angiography had been performed in all patients prior to enrolment into the database (Table 1). None of the patients had CAD or a history of myocardial infarction, systemic arterial hypertension, diabetes mellitus, alcohol abuse, drug dependency, thyroid disease, malignancies, or systemic diseases known to be associated with dilated cardiomyopathy. Exclusion criteria were severe renal failure, liver disease, active infection, allergy, rheumatoid or connective tissue disease, and anti-inflammatory drug treatment. All patients received medical therapy thought to be optimal for their individual CHF profile at that time (Table 2).


View this table:
[in this window]
[in a new window]
  		Table 1 Baseline characteristics of heart failure patients
 

View this table:
[in this window]
[in a new window]
  		Table 2 Laboratory parameters and medication of heart failure patients at study entry
 
The majority of patients was seen at least once per year in our heart failure outpatient clinic. In the remaining patients, annual follow-up was performed by a questionnaire or by phone (Figure 1A). The type of death (sudden cardiac death, cardiac, and non-cardiac death) was determined by two investigators (W.G., M.C.) after an extensive evaluation of the case with the use of informations given by the respective physician and relatives.

Measurement of lipid parameters and sCD14 levels
Blood samples were drawn between 8 a.m. and 9 a.m. after an overnight fast of at least 12 h into sampling tubes (Sarstedt, Nuembrecht, Germany) and immediately transferred to the laboratory for measurements. Total cholesterol and triglyceride measurements of serum samples were done using a commercially available colorimetric system (Roche Diagnostics, Mannheim, Germany). High density lipoprotein (HDL)-cholesterol was determined using a commercially available test using the method of Burstein and Lopes-Virella (Roche Diagnostics). Low density lipoprotein (LDL)-cholesterol levels were calculated using the equation of Friedewald (LDL-cholesterol=fasting cholesterol –1/5 triglyceride–HDL-cholesterol). Apoprotein AI levels were available in 276 patients and were determined using immunologic techniques (ApoA1, Kit Behring Nephelometer 100, Roche Diagnostics). Creatinine clearance was calculated using the formula of Cockcroft and Gault.30

In a subgroup of 82 patients, who have been enrolled during the last years and for whom serum samples of those patients have been stored at –80°C, circulating sCD14 levels were determined using a commercial ELISA system (IBL, Hamburg Germany). sCD14 was immobilized by a polyclonal antibody and sCD14 levels were quantified using a biotin-conjugated second antibody directed against an additional epitope of sCD14 and a peroxidase system.18

Statistical analysis
Results are expressed as median (range) and point estimate (95% confidence intervals). Demographic characteristics are given as mean±standard deviation. The primary objective of the study was to examine whether cholesterol, LDL-cholesterol, HDL-cholesterol, triglycerides, and apolipoprotein I independently contribute to predict adverse prognosis in patients with iDCM. The known influencing factors NYHA functional class, LV-EF, and LVEDD should be included in that analysis.12–15

Univariate and multivariable Cox proportional hazards models were used to evaluate the association between the outcome measures (primary endpoint: time to death or heart transplantation as a combined endpoint; secondary endpoint: time to death alone) and following primary covariates: NYHA functional class, LV-EF, LVEDD, cholesterol, LDL-cholesterol, HDL-cholesterol, triglycerides, and apolipoprotein I. Evaluation of univariate grouped Kaplan–Meier plots did not provide any suggestions for a crude violation of the assumption of proportional hazard rates. For all of these examined variables, which have been included in the models, no information is given that they should be transformed before inclusion into the Cox regression models. It was not planned to compute interactions of the earlier mentioned factors and to include them into the models, because no such interactions have been reported. In addition, there are restrictions concerning the multiplicity of such statistical models and the limited sample size of our database.

Confirmative analysis
According to the primary objective of the study, the five potential factors were tested at a global significance level of 5%. Adjustment for multiple testing was addressed by use of Bonferroni correction. Adjusting for the three known influencing factors, these variables were included in the Cox regression model for the primary endpoint in each of the five primary tests.

It was planned to build prediction models including the known and significant factors from the primary analysis. In further secondary statistical analyses, the value of prognostic factors, which have been discussed in the literature, has been explored.31–34 For building a first Cox regression model, we thereby used a stepwise procedure with an entry level of 0.15. Transplant-free survival probabilities were estimated with the Kaplan–Meier method and the log-rank test was used for comparisons of curves. Calculating with the number of 86 observed events (death or heart transplantation), we are able to differentiate a two-fold increased risk with a power of 80% in two groups with a comparable number of subjects.

Additional statistical analyses such as analysis of variance (ANOVA), linear regression, and Pearson correlation were used for exploration, description, and interpretation. All P-values reported are two-sided and are not adjusted for multiple testing. All statistical calculations were performed using the SAS statistical software package (SAS Institute, Cary, NC, USA).


   Results
Top
 Abstract
 Introduction
 Methods
 Results
Discussion
 Acknowledgement
 References
 
Patient characteristics
Follow-up was defined as the time between measurement of lipid levels and the censoring date or date of event. Median follow-up duration of all patients was 42 months (range 1–105 months). Of the 422 patients, 86 died or received a heart transplant after 1–97 months (median: 24 months) and 71 died after 1–97 months (median: 25 months). The endpoints were as follows: cardiac deaths, n=63 (sudden cardiac deaths, n=28); non-cardiac deaths, n=8; heart transplantation, n=15. The mean follow-up period in 336 Htx-free survivors was 1–105 months (median: 47 months). The cumulative number for death or heart transplantation was 14, 26, and 43 at 6, 12, and 24 months of 422, 402, and 337 patients followed at the respective times. The Kaplan–Meier estimates were 0.974 (0.958–0.989), 0.954 (0.934–0.974), and 0.910 (0.881–0.939) at 6, 12, and 24 months, respectively.

Demographic characteristics and results of electrocardiography and echocardiography of the study patients are given in Table 1. Laboratory measures including lipid levels and the medication of heart failure patients of the study cohort at study entry are displayed in Table 2. At the first follow-up, NYHA functional class has improved in 112 and deteriorated in 100 patients. Cholesterol levels at baseline were 5.8±1.4 mmol/L in patients, in whom NYHA functional class improved, and 5.4±1.3 or 5.1±1.2 mmol/L in patients, in whom NYHA class was stable or deteriorated (ANOVA: P=0.0007; better vs. stable: P=0.02; better vs. worse: P=0.0001; stable vs. worse: P=0.09). Comparable differences were shown when LDL-cholesterol levels were compared with changes of NYHA class (ANOVA: P=0.002; better vs. stable: P=0.01; better vs. worse: P=0.0006; stable vs. worse: P=0.24), whereas no differences were evident for triglycerides, HDL-cholesterol, and apolipoprotein AI (ANOVA: P=0.76, 0.06, and 0.18, respectively). On enrolment, 37 patients were on statins. At the last follow-up, 13 patients were on continuing statin treatment and 10 patients were intermittent users; statins were added in 33 and discontinued in 24 patients.

Cox regression models
Univariate analyses
Cox proportional hazards analyses (Table 3) showed that a higher NYHA functional class, a higher LVEDD, and a lower LV-EF were significant risk factors for death or heart transplantation in iDCM patients. In addition, low cholesterol, HDL-cholesterol, and apolipoprotein AI levels were associated with worse outcome, whereas no significant associations were found for triglyceride and LDL-cholesterol levels. Of interest, the use of beta-blockers and the use of statins were associated with improved outcome. A higher resting heart rate, higher pulmonary capillary wedge and pulmonary artery pressure, and the presence of a left bundle branch block were significant adverse risk factors for the combined endpoint. Age, sex, body mass index, haemoglobin, creatinine, creatinine clearance, potassium, and C-reactive protein levels did not indicate an increased risk for death or heart transplantation in iDCM patients. Comparable results were obtained when mortality alone was used as endpoint (Table 3).


View this table:
[in this window]
[in a new window]
  		Table 3 Univariate Cox proportional hazard analysis for death or heart transplantation and death alone during follow-up
 
Patients were grouped according to their lipid levels (predefined cut-off value: median of total cholesterol, HDL-cholesterol, LDL-cholesterol, or triglyceride levels in the examined population) and log-rank analysis was performed. Transplant-free survival was higher in heart failure patients with cholesterol levels above the median (P=0.03). There was a trend to a lower rate of death or heart transplant in patients with HDL-cholesterol levels above the respective median (P=0.08). No differences were found after patients have been grouped according to LDL-serum levels (P=0.43) or triglyceride levels (P=0.16). In further exploratory analysis, patients were grouped by tertiles of respective lipid data to visualize the dose-response relationship between lipid measurements and outcome (Figure 2). iDCM patients who received statins at any time during the study had a significantly better prognosis than patients who never received cholesterol synthase inhibitors (Figure 3).


Figure 1952
View larger version (33K):
[in this window]
[in a new window]
  		Figure 2 Kaplan–Meier curves showing event-free survival rates for iDCM patients grouped by tertiles of (A) total cholesterol, (B) triglyceride, (C) LDL-cholesterol, and (D) HDL-cholesterol serum levels. P-values refer to the log-rank analysis.

 

Figure 1953
View larger version (17K):
[in this window]
[in a new window]
  		Figure 3 Kaplan–Meier curves showing event-free survival rates for iDCM patients, who received statins at any time during the study or who never used statins. P-values refer to the log-rank analysis.

 
Multivariable analyses
Several multivariable Cox proportional hazards models were analysed (Table 4). After step-wise regression analysis, only LVEDD, NYHA functional class, and LV-EF emerged as parameters independently contributing to the model predicting increased risk for death or heart transplantation. When cholesterol, triglyceride, LDL-cholesterol, HDL-cholesterol, or apolipoprotein AI levels were added to the models, contribution of LVEDD, LV-EF, and NYHA functional class was nearly unchanged and lipid parameters could not be proven to be independent contributors in these models (Table 4; cholesterol: P=0.34; triglycerides: P=0.17; LDL-cholesterol: P=0.84; HDL-cholesterol: P=0.49; apolipoprotein AI: P=0.17). If the patients on statins (Table 2) are removed from statistical analysis, the relationship between cholesterol and transplant free-survival is not essentially altered. The Cox proportional hazards model was recalculated by adding important clinical parameters into the model: serum potassium levels, resting heart rate, pulmonary capillary wedge pressure, pulmonary artery pressure, and the presence of a left bundle branch block did not add any further independent information (Table 5). Comparable results were obtained, when death alone was used as endpoint. A reduced LV-EF did not contribute further independent informations in the multivariable Cox regression models, when death alone was used as endpoint. This may be explained by the some relation of increased LVEDDs and reduced LV-EF. Beta-blocker use was associated with a trend to a better transplant-free survival in iDCM patients and was significantly associated with better outcome when death alone was used as endpoint. Of interest, exploratory multivariable Cox regression models revealed that the use of statins at any time during the study significantly contributed to the model predicting death or heart transplantation or death alone in our population (Table 5).


View this table:
[in this window]
[in a new window]
  		Table 4 Multivariable Cox proportional hazard analysis for death or heart transplantation and death alone during follow-up (lipid parameters)
 

View this table:
[in this window]
[in a new window]
  		Table 5 Multivariable Cox proportional hazard analysis for death or heart transplantation and death alone during follow-up (clinical parameters)
 
Furthermore, we evaluated the contribution of the earlier mentioned parameters to predict adverse outcomes using the exact modes of death as endpoints. Although limited by the small sample sizes, LVEDD, NYHA functional class, and LV-EF emerged as parameters predicting cardiac death (P=0.0005, 0.0061, and 0.07, respectively) or heart transplantation (P=0.017, 0.016, and 0.09, respectively), whereas the use of statins or total cholesterol levels did not add any further important informations. LVEDD and total cholesterol levels independently contributed to the models to predict sudden cardiac death (P=0.049 and 0.048, respectively), while LVEDD and NYHA class did not. None of the variables significantly contributed to the Cox regression models to predict non-cardiac deaths (n=8) due to the small sample size.

Lipid levels and severity of cardiac disease
Associations between total cholesterol levels and NYHA functional class, LVEDD and LV-EF are illustrated in Figure 4. Patients in the NYHA functional class III had lower cholesterol levels than CHF patients in NYHA functional class I (Figure 4A, P=0.03). Total cholesterol levels were negatively associated with the LVEDDs (Figure 4B, P=0.01). In addition, LV-EF was positively associated with cholesterol levels in iDCM patients (Figure 4C, P=0.006). However, both associations were weak. HDL- and LDL-levels were lower in patients with a low LV-EF (LV-EF<16%) when compared with patients with an ejection fraction >25% (HDL: P=0.015; LDL: P=0.008). In contrast, triglycerides were not different among patients with low compared with a high NYHA functional class (NYHA I vs. III: P=0.65). No associations were found for triglyceride levels with LVEDD (P=0.49) and ejection fraction (P=0.23).


Figure 1954
View larger version (29K):
[in this window]
[in a new window]
  		Figure 4 Total cholesterol levels in heart failure patients. Associations of cholesterol levels with NYHA functional class (A; values are mean±SD, NYHA Class: *I vs. III, P=0.03), LVEDD (B), or the LV-EF (C) are displayed.

 
Levels of circulating sCD14 and lipids
Circulating sCD14 levels were examined in a subgroup of 82 heart failure patients with iDCM. Demographic parameters show that this subgroup had an advanced disease compared to the overall group (increased rate of NYHA functional class III, P<0.001; lower LV-EF: P=0.05). Lower levels of sCD14 were associated with lower levels of total cholesterol (r=0.38; P=0.004; Figure 5A) and LDL-cholesterol levels (y=4.62+0.26x; r=0.26; P=0.02). Moreover, reduced sCD14 levels were associated with a low LV-EF (r=0.46; P=0.004; Figure 5B). In contrast, no differences were found when patients were grouped according to NYHA functional class (NYHA I/II vs. III: 5.6±1.1 vs. 5.3±1.1 µg/L, P=0.39).


Figure 1955
View larger version (28K):
[in this window]
[in a new window]
  		Figure 5 Association of total cholesterol (A) and LV-EF (B) with levels of circulating sCD14 levels in iDCM patients (n=82). Total cholesterol levels and LV-EF were associated with circulating sCD14 levels indicating that low sCD14 levels are associated with reduced cholesterol levels and low LV-EF.

 

   Discussion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
Acknowledgement
 References
 
We prospectively evaluated the independent prognostic significance of circulating lipid levels in heart failure patients with iDCM. The main results of the study are as follows. (i) Univariate Cox proportional hazards analysis revealed that low cholesterol and HDL-cholesterol levels are weak risk factors for the combined endpoint ‘death and heart transplantation’ or the endpoint ‘death alone’ in our study population. (ii) High LVEDD, high NYHA functional class, and low LV-EF were confirmed as predictors for death or heart transplantation as assessed by multivariable Cox proportional hazards analysis. Lipid levels did not contribute further independent information. Comparable results were obtained when death alone was evaluated. (iii) Exploratory analyses revealed that variables such as resting heart rate, hypokalaemia, C-reactive peptide and pulmonary artery pressures, and the presence of left bundle branch block did not, whereas the use of beta-blockers or statins did significantly contribute further independent information. (iv) Low levels of circulating CD14 were associated with low cholesterol levels and reduced LV-EF.

An inverse relation of cholesterol levels with the risk of death in either sex has been reported,35 consistent with observations that total cholesterol is negatively associated with non-atherosclerotic death.36–41 Those data have been extended for heart failure patients: low cholesterol levels were reported to predict adverse prognosis in CHF patients with or without CAD8,9,11 even raising theoretical concerns about the use of statins in CHF treatment.7,42 Although low cholesterol levels were weakly associated with poor prognosis in our large population of iDCM patients in univariate analysis, this relation is lost in multivariable analysis. Known risk predictors, such as high NYHA functional class, high LV end-diastolic parameter, and low LV-EF, have been confirmed as independent risk factors predicting time to death and/or heart transplantation. Our data in iDCM patients further suggest that cholesterol levels are negatively associated with the severity of heart failure. Therefore, we speculate that low cholesterol levels in iDCM patients are indicative for the metabolic consequences of the severity of cardiac disease. Such a relation has earlier been reported in patients with cancer41,43 and in critically ill surgical patients.44

The lacking independent contribution of low cholesterol levels in our population to predict mortality in CHF patients may be explained by demographic differences among this and previously reported studies. Although patients with NYHA IV functional class heart failure and cardiac cachexia were included in earlier examinations,8,18,45 patients with NYHA IV functional class were excluded in our study. Of note, cholesterol levels obviously depend on age. An analysis of serum cholesterol levels and mortality risk as a function of age displayed that the relationship between total cholesterol levels and all-cause mortality was positive at the age of 40 years, negligible at 50–70 years, and negative at the age of 80 years.46 Our study population was younger than the patients examined before.8,18,45 Thus, the independent contribution of low cholesterol levels in CHF patients in those analyses may be explained by unrecognized non-cardiac disease in elderly patients.43

It was proposed that high cholesterol levels beneficially modulate inflammatory activation by neutralization of endotoxin,7,8,18 and circulating sCD14 is seen as a marker of endotoxin challenge. In our study, low sCD14 levels are associated with a reduced LV-EF and low cholesterol levels. Because low levels of sCD14 may be due to a reduced shedding/formation of sCD14 by macrophages, we speculate that the metabolic consequences of severe heart failure initiate systemic immunosuppression.9 Of note, a significant positive relation of inflammatory markers and lipoprotein levels were found in CAD patients, but no associations were found in iDCM patients.47 Therefore, conclusions drawn from mixed heart failure population including the suggested ‘endotoxin-lipoprotein hypothesis’7 have to be interpreted with caution.

Intake of lipid lowering drugs appears mandatory in CAD patients. However, prospective data in patients with severe heart failure with or without CAD are currently scarce. It has even been postulated that decreasing lipid levels by statin treatment may adversely affect heart failure.7,8 However, experimental and clinical evidence supports the use of those drugs in CHF patients. Cerivastatin treatment of rats with experimental heart failure beneficially modulates LV remodelling and function.48 The short-term treatment of 68 iDCM patients with simvastatin significantly improved LV function, clinical symptoms, and markers of inflammatory activation.42 Comparable results have been reported in a small trial examining the effect of cerivastatin in iDCM patients.49 Possible beneficial effects of statins in CHF patients are also supported by the exploratory analysis of our database: statin use was associated with improved event-free survival in iDCM patients. Thus, statin treatment in heart failure patients appears promising. However, recommendations cannot be given until final results of ongoing prospective intervention trials are available.

There are several limitations of this study. (i) Enrolment of consecutive patients was not homogeneous throughout the years, which is probable due to changes of referral to our centre. Furthermore, lipid data were lacking in several patients leading to the exclusion of those patients for further analysis, which may have reduced external validity. (ii) The patients in our database displayed a very good prognosis indicating that we may have examined a rather odd population of patients. However, recent large intervention trials report comparable findings on prognosis in patients with iDCM.50,51 (iii) The Marburg cardiomyopathy database is a prospective database in which all deaths or heart transplants are included as primary endpoints. Because elective heart transplantation may be influenced by the size and blood group of the patients, our results may be distorted using this ‘soft endpoint’. Nevertheless, data analysis using ‘death alone’ as secondary endpoint did not largely change our results. (iv) In previously published populations, higher age or increased body mass index are related to adverse outcome. Those co-morbidities may limit survival by influencing the progression of CAD and/or hypertensive heart disease. Because those co-morbidities were not evident in our cohort, conclusions to the overall heart failure population should be drawn carefully. (v) Multiple testing may lead to associations by chance leading to a final model deeming unreliable. Therefore, only a limited number of primary variables were included in the primary analysis and the Bonferroni method was used to correct for multiple testing. Our results clearly show that none of the candidate factors appears promising to independently predict prognosis in patients with iDCM. In conclusion, our statistical analyses have not led to the inclusion of additional factors in the final Cox regression model and, thus, confirm previous analyses.12–15

In summary, our findings in iDCM patients corroborate and extend previous findings that low cholesterol levels indicate worse prognosis. However, these associations were weak and multivariable Cox proportional hazards analysis revealed that only NYHA functional class, LVEDD, and LV systolic ejection fraction are independent predictors for adverse outcome in this population. Our study suggests that low cholesterol levels are rather indicators of severe heart failure. This results and findings of small clinical studies do not argue against the use of statins in heart failure patients.


   Acknowledgement
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Acknowledgement
References
 
We thank Marlies Krombach for her expert technical assistance. No financial interests have to be declared.


   References
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Acknowledgement
 References
 

  1. Ross R. Atherosclerosis—an inflammatory disease. N Engl J Med 1999;340:115–126.[Free Full Text]
  2. Cannon CP, Braunwald E, McCabe CH et al. Intensive versus moderate lipid lowering with statins after acute coronary syndromes. N Engl J Med 2004;350:1495–1504.[Abstract/Free Full Text]
  3. Nadar S, Lim HS, Beevers DG et al. Lipid lowering in hypertension and heart protection: observations from the Anglo-Scandinavian Cardiac Outcomes Trial (ASCOT) and the Heart Protection Study. J Hum Hypertens 2002;16:815–817.[CrossRef][ISI][Medline]
  4. Scandinavian Simvastatin Survival Study Groups Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 1994;344:1383–1389.[CrossRef][ISI][Medline]
  5. The Long-Term Intervention with Pravastatin in Ischaemic Disease (LIPID) Study Group. Prevention of cardiovascular events and death with pravastatin in patients with coronary heart disease and a broad range of initial cholesterol levels. N Engl J Med 1998;339:1349–1357.[Abstract/Free Full Text]
  6. Kjekshus J, Pedersen TR, Olsson AG et al. The effects of simvastatin on the incidence of heart failure in patients with coronary heart disease. J Card Fail 1997;3:249–254.[CrossRef][Medline]
  7. Rauchhaus M, Coats AJ, Anker SD. The endotoxin-lipoprotein hypothesis. Lancet 2000;356:930–933.[CrossRef][ISI][Medline]
  8. Rauchhaus M, Clark AL, Doehner W et al. The relationship between cholesterol and survival in patients with chronic heart failure. J Am Coll Cardiol 2003;42:1933–1940.[Abstract/Free Full Text]
  9. Vredevoe DL, Woo MA, Doering LV et al. Skin test anergy in advanced heart failure secondary to either ischemic or idiopathic dilated cardiomyopathy. Am J Cardiol 1998;82:323–328.[CrossRef][ISI][Medline]
  10. Richartz BM, Radovancevic B, Frazier OH et al. Low serum cholesterol levels predict high perioperative mortality in patients supported by a left-ventricular assist system. Cardiology 1998;89:184–188.[CrossRef][ISI][Medline]
  11. Horwich TB, Hamilton MA, Maclellan WR et al. Low serum total cholesterol is associated with marked increase in mortality in advanced heart failure. J Card Fail 2002;8:216–224.[CrossRef][ISI][Medline]
  12. Neglia D, Michelassi C, Trivieri MG et al. Prognostic role of myocardial blood flow impairment in idiopathic left ventricular dysfunction. Circulation 2002;105:186–193.[Abstract/Free Full Text]
  13. Grzybowski J, Bilinska ZT, Ruzyllo W et al. Determinants of prognosis in nonischemic dilated cardiomyopathy. J Card Fail 1996;2:77–85.[CrossRef][Medline]
  14. Komajda M, Jais JP, Reeves F et al. Factors predicting mortality in idiopathic dilated cardiomyopathy. Eur Heart J 1990;11:824–831.[Abstract/Free Full Text]
  15. Grimm W, Christ M, Bach J et al. Noninvasive arrhythmia risk stratification in idiopathic dilated cardiomyopathy: results of the Marburg Cardiomyopathy Study. Circulation 2003;108:2883–2891.[CrossRef][ISI][Medline]
  16. Cohn JN, Ferrari R, Sharpe N. Cardiac remodeling—concepts and clinical implications: a consensus paper from an international forum on cardiac remodeling. Behalf of an International Forum on Cardiac Remodeling. J Am Coll Cardiol 2000;35:569–582.[Abstract/Free Full Text]
  17. Brunner-La Rocca HP, Esler MD, Jennings GL et al. Effect of cardiac sympathetic nervous activity on mode of death in congestive heart failure. Eur Heart J 2001;22:1136–1143.[Abstract/Free Full Text]
  18. Anker SD, Egerer KR, Volk HD et al. Elevated soluble CD14 receptors and altered cytokines in chronic heart failure. Am J Cardiol 1997;79:1426–1430.[CrossRef][ISI][Medline]
  19. Niebauer J, Volk HD, Kemp M et al. Endotoxin and immune activation in chronic heart failure: a prospective cohort study. Lancet 1999;353:1838–1842.[CrossRef][ISI][Medline]
  20. Flegel WA, Baumstark MW, Weinstock C et al. Prevention of endotoxin-induced monokine release by human low- and high-density lipoproteins and by apolipoprotein AI. Infect Immun 1993;61:5140–5146.[Abstract/Free Full Text]
  21. Harris HW, Grunfeld C, Feingold KR et al. Human very low density lipoproteins and chylomicrons can protect against endotoxin-induced death in mice. J Clin Invest 1990;86:696–702.[ISI][Medline]
  22. Levine DM, Parker TS, Donnelly TM et al. In vivo protection against endotoxin by plasma high density lipoprotein. Proc Natl Acad Sci USA 1993;90:12040–12044.[Abstract/Free Full Text]
  23. Netea MG, de Bont N, Demacker PN et al. Lipoprotein(a) inhibits lipopolysaccharide-induced tumor necrosis factor alpha production by human mononuclear cells. Infect Immun 1998;66:2365–2367.[Abstract/Free Full Text]
  24. Wurfel MM, Kunitake ST, Lichenstein H et al. Lipopolysaccharide (LPS)-binding protein is carried on lipoproteins and acts as a cofactor in the neutralization of LPS. J Exp Med 1994;180:1025–1035.[Abstract/Free Full Text]
  25. Anker SD, Ponikowski PP, Clark AL et al. Cytokines and neurohormones relating to body composition alterations in the wasting syndrome of chronic heart failure. Eur Heart J 1999;20:683–693.[Abstract/Free Full Text]
  26. Rauchhaus M, Doehner W, Francis DP et al. Plasma cytokine parameters and mortality in patients with chronic heart failure. Circulation 2000;102:3060–3067.[Abstract/Free Full Text]
  27. Funda DP. CD14 is expressed and released as soluble CD14 by human intestinal. Infect Immun 2001;69:3772–3781.[Abstract/Free Full Text]
  28. Ziegler-Heitbrock HW, Ulevitch RJ. CD14: cell surface receptor and differentiation marker. Immunol Today 1993;14:121–125.[CrossRef][ISI][Medline]
  29. Schiller NB, Shah PM, Crawford M et al. Recommendations for quantitation of the left ventricle by two-dimensional echocardiography. American Society of Echocardiography Committee on Standards, Subcommittee on Quantitation of Two-Dimensional Echocardiograms. J Am Soc Echocardiogr 1989;2:358–367.[Medline]
  30. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976;16:31–41.[ISI][Medline]
  31. Smilde TD, Hillege HL, Navis G et al. Impaired renal function in patients with ischemic and nonischemic chronic heart failure: association with neurohormonal activation and survival. Am Heart J 2004;148:165–172.[CrossRef][ISI][Medline]
  32. Laragh JH, Sealey JE. K+ depletion and the progression of hypertensive disease or heart failure. The pathogenic role of diuretic-induced aldosterone secretion. Hypertension 2001;37:806–810.[Free Full Text]
  33. Kaneko K, Kanda T, Yamauchi Y et al. C-Reactive protein in dilated cardiomyopathy. Cardiology 1999;91:215–219.[CrossRef][ISI][Medline]
  34. van der Meer P, Voors AA, Lipsic E et al. Prognostic value of plasma erythropoietin on mortality in patients with chronic heart failure. J Am Coll Cardiol 2004;44:63–67.[Abstract/Free Full Text]
  35. Gordon T, Kannel WB, Castelli WP et al. Lipoproteins, cardiovascular disease, and death. The Framingham study. Arch Intern Med 1981;141:1128–1131.[Abstract]
  36. Ducimetiere P, Eschwege E, Richard J et al. Relationship of glucose tolerance to prevalence of ECG abnormalities and to annual mortality from cardiovascular disease: results of the Paris Prospective Study. J Chronic Dis 1979;32:759–766.[CrossRef][ISI][Medline]
  37. Hawthorne VM, Gilmour WH. Relationship of glucose to prevalence of ECG abnormalities at baseline and to 6-yr mortality in Scottish males aged 45–64 yr. J Chronic Dis 1979;32:787–796.[CrossRef][ISI][Medline]
  38. Beaglehole R, Foulkes MA, Prior IA et al. Cholesterol and mortality in New Zealand Maoris. Br Med J 1980;280:285–287.[ISI][Medline]
  39. Law MR. Serum cholesterol and cancer. Br J Cancer 1992;65:307–308.[ISI][Medline]
  40. Epstein FH. Low serum cholesterol, cancer and other noncardiovascular disorders. Atherosclerosis 1992;94:1–12.[CrossRef][ISI][Medline]
  41. Song YM, Sung J, Kim JS. Which cholesterol level is related to the lowest mortality in a population with low mean cholesterol level: a 6.4-year follow-up study of 482,472 Korean men. Am J Epidemiol 2000;151:739–747.[Abstract/Free Full Text]
  42. Node K, Fujita M, Kitakaze M et al. Short-term statin therapy improves cardiac function and symptoms in patients with idiopathic dilated cardiomyopathy. Circulation 2003;108:839–843.[CrossRef][ISI][Medline]
  43. Rose G, Shipley MJ. Plasma lipids and mortality: a source of error. Lancet 1980;1:523–526.[ISI][Medline]
  44. Gui D, Spada PL, De Gaetano A et al. Hypocholesterolemia and risk of death in the critically ill surgical patient. Intensive Care Med 1996;22:790–794.[ISI][Medline]
  45. Anker SD, Cicoira M. Chronic heart failure and cardiac cachexia and links between the endocrine and immune systems. Z Kardiol 1999;88 (Suppl. 3):S18–S23.
  46. Kronmal RA, Cain KC, Ye Z et al. The analysis of serum cholesterol levels and mortality risk as a function of age. A report based on the Framingham data. Arch Intern Med 1993;153:1065–1073.[Abstract]
  47. Conraads VM, Bosmans JM, Schuerwegh AJ et al. Association of lipoproteins with cytokines and cytokine receptors in heart failure patients. Differences between ischaemic versus idiopathic cardiomyopathy. Eur Heart J 2003;24:2221–2226.[Abstract/Free Full Text]
  48. Bauersachs J, Galuppo P, Fraccarollo D et al. Improvement of left ventricular remodeling and function by hydroxymethylglutaryl coenzyme a reductase inhibition with cerivastatin in rats with heart failure after myocardial infarction. Circulation 2001;104:982–985.[Abstract/Free Full Text]
  49. Laufs U, Wassmann S, Schackmann S et al. Beneficial effects of statins in patients with non-ischemic heart failure. Z Kardiol 2004;93:103–108.[CrossRef][ISI][Medline]
  50. Bansch D, Antz M, Boczor S et al. Primary prevention of sudden cardiac death in idiopathic dilated cardiomyopathy: the Cardiomyopathy Trial (CAT). Circulation 2002;105:1453–1458.[Abstract/Free Full Text]
  51. Kadish A, Dyer A, Daubert JP et al. Prophylactic defibrillator implantation in patients with nonischemic dilated cardiomyopathy. N Engl J Med 2004;350:2151–2158.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?

Related articles in EHJ:

Threats, opportunities, and statins in the modern management of heart failure
John G.F. Cleland, Huan Loh, Jonathan Windram, Kevin Goode, and Andrew L. Clark
EHJ 2006 27: 641-643. [Extract] [Full Text]  



This article has been cited by other articles:


Home page
 Eur Heart JHome page
J. G.F. Cleland, H. Loh, J. Windram, K. Goode, and A. L. Clark
Threats, opportunities, and statins in the modern management of heart failure
Eur. Heart J., March 2, 2006; 27(6): 641 - 643.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
27/6/691    most recent
ehi195v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Related articles in EHJ
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (10)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Christ, M.
Right arrow Articles by Maisch, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Christ, M.
Right arrow Articles by Maisch, B.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?