LIFESTAT - Living With Statins, a Cross Sectional Study

Background

HYPERCHOLESTEROLEMIA AND STATIN USE IN DENMARK

Simvastatin is the most commonly prescribed statin, a class of drugs that inhibit hydroxyl-methyl-glutaryl (HMG) coenzyme A reductase, and thereby blocking biosynthesis of cholesterol in the liver. Simvastatin is prescribed for individuals with elevated low-density lipoprotein cholesterol (LDL-C) and/or total cholesterol, because these clinical parameters are viewed as a risk factor for cardiovascular-disease (CVD), even in the absence of other health problems or risk factors, such as previous myocardial infarction, diabetes or hypertension.

Approximately 40% of the prescriptions for statins are issued for primary prevention of elevated cholesterol by general practitioners to patients without bodily symptoms or signs. Only the "cholesterol number" makes the risk of heart attack and stroke visible. The lack of symptoms is likely to be of importance for patients' adherence to treatment as is adverse effects. A number of factors, such as information in mass media and changes in daily life, may affect the decision to take the treatment.

TREATMENT GUIDELINES FOR HYPERCHOLESTEROLEMIA

The guidelines (10; 12) indicate preventive treatment with statins is appropriate in individuals with >10% predicted risk of a major vascular event within 5 years, while, some, but not all opinion-leaders advocate a 5% threshold (3; 10). Nevertheless, statin therapy failed to reduce all-cause mortality in a meta-study of 65,229 patients without CVD, some of whom had diabetes (13). Similarly, a Cochrane review analysis, which included some studies in which more than 10% of the patients had history of CVD, showed only 0.5% reduction in all-cause mortality, indicating that for every 200 patients taking statins daily for 5 years, 1 death would be prevented (15). These data suggest that more conservative use of statins to prevent CVD in otherwise healthy individuals at low risk for future CVD may be warranted.

THE DOWN-SIDE

Rhabdomyolysis (skeletal muscle cell death) is an infrequent but serious side-effect of statin use, that can on rare occasion lead to acute renal failure and death (i.e., 1.5 deaths per 106 prescriptions (11)). Statin use is much more frequently associated with muscle dysfunction, including myalgia (muscle pain), cramps, and weakness. The reported incidence of myalgia varies from 1% (pharmaceutical company reports) to as high as 75% in statin-treated athletes (9; 11). Mild to severe myalgia is a strong disincentive to regular exercise, and because regular exercise is one of the critical life-style approaches to preventing CVD and reducing blood cholesterol, this is a significant down-side of statin use. Regular exercise is also effective in preventing and treating obesity and type 2 diabetes, which themselves are risk factors for CVD (18).

The mechanism behind the myalgia is not known. However, we have recently demonstrated that muscle mitochondrial function is impaired with statin treatment and the Q10 protein may play a key role in this (8). In addition, the statins also negatively affect the glucose tolerance (8), increasing the risk of type 2 diabetes.

RESEARCH QUESTIONS:

The overarching research question is: why does statin treatment cause muscle pain? We are not the only research group in the world that try to answer that question, but we are the only one that has indeed provided a mechanistic explanation, and provided a proof-of-concept (8). We will now test this in a larger patient population.

Our background in muscle and exercise physiology and in bioenergetics makes it natural to further ask:

a Does statin treatment impair (or even prohibit) physical exercise training?

b Does statin treatment cause:

• Decreased muscle strength?
• Skeletal muscle inflammation?
• Decreased mitochondrial respiratory function? c Abnormal glucose homeostasis?

Re question b \& c: If so, can physical training counteract this effect of statin treatment?

Methodology

COHORT

Patients that fulfil defined inclusion and exclusion criteria will be recruited from pharmacies and through advertisement in newspapers in Copenhagen. The vast majority of these patients are being treated on basis of the HeartScore risk estimation system that offers direct estimation of the ten-year risk of fatal cardiovascular disease in a format suited to the constraints of clinical practice (16) (www.HeartScore.org). A staggered recruitment will be implemented.

60 patients both men and women (age: 40-70 years; BMI: 25-35 kg/m2) that are being treated with Simvastatin minimum 40 mg/day as primary prevention are recruited. No other risk factors for CVD except elevated total cholesterol and/or elevated LDL cholesterol and mild hypertension (<145/100 mm Hg) must be present.

The test subjects will undergo three experimental days as described below:

Experimental days:

Day 1 (½ day - overnight fasting):

• Medical history; clinical examination + ECG. Measurements: Blood pressure, weight, height, W/H-ratio, thigh circumference.
• Dual Energy X-ray Absorptiometry-scan (body composition and body fat).
• FatMax test and maximal oxygen uptake test (VO2-max) or ergometer bike.

Day 2 (½ day - overnight fasting):

• Oral glucose tolerance test + score questionnaire for muscle pain/discomfort (incl. Visual Analog Scale)
• Isokinetic strength and Rate of Force Development (PowerRig and KinCom dynamometer).
• Repeated VO2-max-test

Day 3 (1 day - overnight fasting):

• Muscle biopsy, vastus lateralis (before and after clamp) and fat biopsy from the subcutaneous adipose tissue in the abdomen.
• Intravenous glucose tolerance test (IVGTT)
• Euglycemic, hyperinsulinaemic clamp.

Statistical considerations

The major end-points are all end-point which we have tested before in other clinical populations. In general, in order to detect a 10% difference in these parameters before vs. after a training program or between statin users and control, requires 20-25 subjects in each group if an alpha level of <0.05 and risk of type 2 error is set to 10%. 20-25 subjects are necessary if the "conventional" 20% type 2 error risk is implemented. Thus, the present study has a considerable safety-margin in terms of statistical power.