Clinical Drug Experience Knowledgebase

A Study Of The Usage Of Statins In A Community Heart Failure Population: The Impact of HMG Co-Enzyme A Reductase Inhibitors on Heart Structural Parameters, Brain Naturetic Peptide, Markers of Inflammation and Fibrosis

Heart Failure (HF) remains a major public health problem. Despite significant advances in the pharmacological management of this condition, the high mortality associated with heart failure remains a concern with annual figures ranging from 10 - 15% in mildly symptomatic patients to 50 - 60% in patients with severe heart failure.1

The major aetiological cause of HF is ischemic heart disease (IHD). While guidelines on the treatment of IHD recommend the use of lipid lowering therapy with hydroxymethylglutaryl-coenzyme A reductase inhibitors (statins) to improve prognosis, recent surveys from the UK and have shown that these effective therapies are being under-utilised.2 The reasons appear to be multifactorial but include issues related to efficacy, safety, potential for adverse drug reactions, failure to prescribe appropriate medication or dose and noncompliance with therapy.3

Some data from the large trials of statins are suggestive of benefit. In the 4S study 412 patients (9.2% of the study population) developed chronic heart failure requiring treatment, i.e. 228 (10.3%) in the placebo group and 184 (8.3%) patients in the group treated with simvastatin (p < 0.015). However, several questions remain, not least how statins might be beneficial in the setting of symptomatic heart failure.4 Treatment of heart failure has been targeted to a certain extent at the cause - and in the case of the majority of heart failure patients, the cause is related to ischemic heart disease and, perhaps, atherosclerosis. Furthermore, it is believed that the non-lipid lowering effects of statins, including neoangiogenesis, inhibition of proinflammatory cytokine activity, anti-fibrotic effects and favorable modulation of the autonomic nervous system, could play a positive role in the medical management of HF.5,6

However, there are several specific concerns, which clinicians may have in relation to the use of statins in heart failure. The average age of a community heart failure population is generally above 70 years and observational data from Framingham have suggested an inverse relationship between total cholesterol levels and all-cause mortality in older patients.5 There is an association between a total cholesterol level less than 5.2 mmol/L and impaired one-year event free survival in a large group of patients with heart failure.6 Statins can have a deleterious effect on myocardial function and thus have adverse effects on skeletal or cardiac muscles due to the inhibition of the synthesis of mevalonate, a precursor of ubiquinone.7

Elevated levels of markers such as C-reactive protein, TNF alpha and other inflammatory cytokines have been associated with poorer outcome in patients with HF, but excluding those with clear sources of infection or other inflammatory disorder, elevated C-reactive protein correlates with NYHA functional class and is associated with higher levels of morbidity.9 Several therapies in HF, notably ACE inhibition and AII Receptor Blockade have been shown to have anti-cytokine effects.10 Statins have been shown to reduce levels of circulating inflammatory cytokines in IHD. Recent data using atorvastatin has suggested that significant reductions in C-reactive protein levels are only seen in the patients with the highest pre-existing levels.11

Chronic inflammation can promote excessive cardiomyocyte loss, cardiac fibroblast proliferation, increased rates of extracellular matrix (ECM) turnover and altered composition of the ECM resulting in reduced functional performance of the heart.12-17

Natriuretic peptides have been shown to be useful markers for screening, diagnosis and treatment monitoring in heart failure. New data suggests that BNP may be an important counter-regulatory protein operating through the natriuretic peptide receptor type-A (NPRA) as a protective response, not only to stretch and volume overload in the myocardium13 but also to myocardial fibrosis and inflammation.18

In summary, from a clinical practice point of view the under usage of statins in heart failure populations may reflect the ongoing debate about their efficacy, safety and the potential for adverse drug interactions in this patient population. From a mechanistic point of view, the potential benefits of statins in heart failure may relate to direct improvements in ischemia and/or cardiac function and may be mediated via suppression of inflammatory cytokines, alteration of myocardial fibrosis and may be unrelated to cholesterol lowering effects. This project will have two phases with the following objectives:

[Objective 1] To evaluate the usage of statins in a community heart failure population admitted to our institution.

[Objective 2] To study the impact of statin therapy on natriuretic peptides, inflammatory markers and markers of fibrosis in a population with heart failure and normal cholesterol

To evaluate objective 2, a prospective, randomized, open-label study of mild to moderate HF patients was approved by the St Vincent's University Hospital ethics committee. All recruited patients give their informed consent and are randomized using a computer generated protocol to atorvastatin therapy [Intervention group (I)] or no statin therapy [Control group (C)] for a six month period.

Additionally, a further group of individuals without heart failure or proven cardiovascular disease and with normal cholesterol are enrolled for the purposes of baseline comparison (Normal group).

In the randomized study, all patients have mild to moderate HF (NYHA class II-III) of any aetiology and documented left ventricular systolic dysfunction (LVEF < 45%) by echocardiography within 3 months of randomization. These patients are clinically stable, on optimal HF medical therapy which had remained unaltered for at least four weeks prior to recruitment. All have normal fasting total cholesterol (TC < 5.1 mmol/L). The following patients are excluded from the study: those receiving lipid lowering agents; those with absolute or relative contraindications to statins; those with known chronic inflammatory conditions; patients with medical conditions requiring anti-inflammatory or immuno-suppressive therapies. The same exclusion criteria are applied to the Normal group.

In addition to optimal HF therapy, the I group (n=28) receive atorvastatin titrated from 10-40 mg/day over 3 months and maintained at 40mg/day for a further 3 months. The C group (n=28) receive optimal HF medication only.

Patients undergo routine clinical examination at the heart failure unit, which includes clinical history, physical examination and chest X-Ray appearance. Fasting Blood samples are taken and stored for subsequent measurement of cholesterol, FBC, High Sensitivity CRP, TNF alpha, IL-6, U & E, TFT, BNP and markers of collagen turnover. 12-lead ECG and Echocardiography is performed at baseline and six months. The Minnesota Living With Heart Failure (MLWHF) questionnaire was self-administered at baseline and 6 months. The Normal group underwent phlebotomy at baseline to enable comparison with the HF study sample.

Routine clinical review at the heart failure unit is carried out during the trial, including clinical examination FBC (including LFTs), U & E and assessment for side effects. The patients are advised to attend their general practitioner for all non heart failure related problems.

Comparisons between the Normal group and the randomized study patients (C and I groups) at baseline are made using independent t-test, Mann Whitney or chi square test where appropriate. Due to non normal distributions in the biochemical markers and lipids, Friedman's ANOVA is used to test for treatment effects over the study period. Pairwise comparisons are made using the Wilcoxon test. Repeated measures analysis of variance (RM ANOVA) is used to test treatment effects on quality of life (QOL) and LVEF. To enable statistical control for the effects of age and gender, non normal variables are log transformed and a RM ANCOVA fit to the data. Correlations are computed using Spearman's rank order correlation.