REVIEW URRENT C OPINION
The importance of considering LDL cholesterol response as well as cardiovascular risk in deciding who can benefit from statin therapy Handrean Soran a,b, Jonathan D. Schofield a,b, and Paul N. Durrington a
Purpose of review Guidelines seeking to deploy statin treatment rely heavily on the use of estimates of absolute cardiovascular disease (CVD) risk as an arbiter of who should receive statins. We question whether this is an effective strategy unless the LDL-cholesterol (LDL-C) response is also considered. Recent findings Recently, meta-analyses of randomized clinical trials of statins have revealed that CVD risk decreases linearly by 22% for each 1 mmol/l reduction in LDL-C. Calculation of the number needed to treat with statins to prevent one CVD event using both the pretreatment absolute CVD risk and the LDL-C response that can be achieved is thus possible. Application of this evidence reveals that many people (including younger ones) with high LDL-C levels can benefit more than people currently receiving statin treatment solely on the basis of their absolute CVD risk, whereas others at higher CVD risk, but with lower LDL-C, will derive little benefit. This does not seem to have been adequately considered in recent clinical guidelines. Summary A simple additional mathematical step in risk assessment to take account of the LDL-C response to statins and provide knowledge of number needed to treat would greatly improve individual management, costeffectiveness and the population impact of statins. Keywords absolute CVD risk, LDL cholesterol, number needed to treat, population impact of statin therapy
INTRODUCTION In a meta-analysis of 26 randomized clinical trials involving 170 000 participants, statins have been shown to decrease atherosclerotic cardiovascular disease (CVD) morbidity and mortality [1]. In this meta-analysis, the definition of CVD included coronary heart disease and cerebrovascular clinical events (both fatal and nonfatal), the primary end point of most of the trials. The typical reduction in LDL-cholesterol (LDL-C) in trials comparing active statin treatment with placebo was 1.07 mmol/l. In many of these trials, less potent statins were used than were later developed. Even so, the overall decrease in the rate of new CVD events (coronary heart disease and stroke) was 24%, which was highly statistically significant [1,2]. This 24% risk reduction is a relative risk reduction in the sense that it compares the event rate in the statin-treated (actively treated) trial participants with the event rate in those receiving placebo. The absolute number of events prevented during a period of treatment with statins (in other words the actual benefit) will
depend on the event rate in the controls receiving placebo (the rate that would have occurred had statin not been administered). This figure is frequently expressed in guidelines as the number of CVD events occurring in 100 people over the next 10 years, often termed the percentage (%) 10-year risk. Thus, statin treatment in a group of people whose absolute CVD risk is 20% (i.e. 20 per 100) over the next 10 years would be expected to decrease the anticipated rate by 20 0.24%, thus preventing
a Cardiovascular Research Group, School of Biomedicine, University of Manchester, and bCardiovascular Trials Unit, University Department of Medicine, Central Manchester University Hospitals NHS Foundation Trust, Manchester, UK
Correspondence to Professor Paul N. Durrington Cardiovascular Research Group, School of Biomedicine, Core Technology Facility (3rd Floor), University of Manchester, 46 Grafton Street, ManchesterM13 9NT, UK. Tel: +44 161 275 1201; fax: +44 161 275 1183; e-mail: [email protected] Curr Opin Lipidol 2014, 25:239–246 DOI:10.1097/MOL.0000000000000097
0957-9672 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-lipidology.com
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Hyperlipidaemia and cardiovascular disease
KEY POINTS Many people whose absolute CVD risk is considered too low for statin therapy would benefit more than those at higher risk, but with lower LDL-C. Number needed to treat (NNT) is a better way of assessing who should receive statin treatment than simply absolute CVD risk. CVD risk decreases linearly by 22% for each 1 mmol/l reduction in LDL-C achieved with statin treatment. NNT for 10 years to prevent one CVD event ¼ 100 (absolute CVD risk LDL-C decrease 0.22), in which absolute CVD risk is the percentage over the next 10 years and LDL-C decreases in millimoles per litre (multiply NNT by 39, if milligrams per decilitre is the unit used for cholesterol).
4.8 events per 100 people treated. The number of people who must receive treatment for 10 years to prevent one event is calculated by dividing 100 by 4.8, which, in this example, approximates to 21. This figure is referred to as the NNT. When the quality of life of those spared with a CVD event is unimpaired, as it is in the case of primary prevention of CVD with a statin, the NNT and, of course, the cost of treatment itself are the major determinants of the cost of a quality-adjusted life year (QALY) [3]. The relative CVD risk reduction with statin treatment (expressed as the ratio between the rate in treated participants to that in the controls receiving placebo or less efficacious statin treatment in terms of LDL-C lowering) is not constant and varies with the degree to which LDL-C concentration is decreased [1]. The statin trials overall reveal that relative CVD risk declines linearly with respect to the therapeutically induced reduction in LDL-C [1,4]. Meta-analysis further reveals that the percentage decrease in CVD is 22% (95% CI 20, 24) for each 1 mmol/l reduction in LDL-C [1]. The decreased CVD incidence with statins relates most strongly to effects on LDL-C [5] and, although statins differ as to their action on HDL-C levels, this is inconsequential [6 ]. The linear decrease of 22% in relative CVD incidence for each 1 mmol/l reduction in LDL-C does seem to persist down to statin-induced levels of 2 mmol/l and possibly lower [1]. Furthermore, it also applies regardless of the major source of CVD risk, be it diabetes, high blood pressure, smoking, low HDL-C, raised triglycerides or previous CVD. Statins, such as atorvastatin and rosuvastatin, can achieve decreases in LDL-C of more than 50% [7]. International and national clinical recommendations for statin treatment have frequently not taken into account the degree of LDL-C reduction that can be
achieved and have focussed very much on absolute CVD risk as an indication for statin therapy. Misleading conclusions have been reached, particularly in primary prevention, which have major consequences for optimal cost-effective statin deployment in the population and the treatment of individuals.
CURRENT AND PROPOSED RECOMMENDATIONS FOR STATIN TREATMENT The value of statin treatment to reduce CVD is not disputed, but current national and international guidelines focus on baseline CVD risk as the main criterion for prescribing statins [8–11]. Because most CVD events occur in people at average CVD risk simply because they are the most numerous (Rose effect) [12], it has been proposed both by the American College of Cardiology and American Heart Association (ACC/AHA) and by the National Institute for Health and Care Excellence (NICE) that statin use should no longer be confined to particularly high-risk individuals, but extended to people whose risk, at least from middle age onwards, is little more than the population average [13 ,14]. The evidence has still largely been neglected that the benefits of statin therapy relate to the extent of LDL-C reduction, which is greater in those with higher LDL-C and with more potent statins. This review considers a method of taking into account the extent of LDL-C lowering which can be achieved in individual patients, which, combined with the absolute CVD risk, permits the calculation of NNT and provides a more discriminating means of determining who should receive statin treatment. We are led to conclude that a simple formula to modify current estimates of absolute CVD risk estimation to provide NNT would greatly improve clinical guidance, particularly in primary prevention, by more accurately identifying people who can derive substantial benefit, including some from sections of the population at lower absolute risk, in whom more events can be prevented, than in many higher risk patients with lower LDL-C levels. &
&&
240
www.co-lipidology.com
CALCULATION OF NUMBER NEEDED TO TREAT The formula for calculating NNT to prevent one CVD event with statin therapy, taking into account the decrease in LDL-C achieved, is derived as follows: Events prevented per 100 people (e) ¼ Absolute CVD risk LDL-C decrease 0.22 NNT ¼ 100 e NNT ¼ 100 ðabsolute CVD risk LDL cholesterol decrease 0:22Þ % over next 10 years: mmol=l: Volume 25 Number 4 August 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
LDL response as a statin indication Soran et al.
APPLICATION OF NUMBER NEEDED TO TREAT TO STATIN INDICATIONS To demonstrate the use of the equation to calculate the statin NNT in primary prevention, taking into account the extent of LDL-C lowering, let us calculate the NNT for someone with no history of CVD who has an absolute CVD risk of 20% over the next 10 years. If the pretreatment LDL-C is, say, 3.5 mmol/l, decreasing this by 1 mmol/l to 2.5 mmol/l will prevent 4.4 CVD events per 100 similar people treated (e ¼ 20 1 0.22), meaning the number needed to receive treatment to prevent one event is 23, that is, the NNT is 23 (100 4.4). Suppose we now use this formula to calculate the NNT for someone at a 10 year CVD risk of 10% whose pretreatment LDL-C is 4.5 mmol/l. Achieving an LDL-C goal of 2.5 mmol/l interestingly also results in an NNT of 23 (100 (10 2 0.22) ¼ 23). This patient, however, would not, for example, have received statin treatment according to current NICE guidance, which states that patients must have a 10-year CVD risk of 20% or more to be treated [8]. If the NNT is calculated for someone with a pretreatment LDL-C of 5.5 mmol/l, reaching a target LDL-C of 2.5 mmol/l has an NNT value of 22, similar to the
(a)
earlier two examples, even when the 10-year CVD risk is as low as 7%. The degree to which LDL-C can be lowered should therefore be taken into account, as well as the CVD risk, in deciding who should receive statin therapy. Figure 1a shows the NNT to prevent one event in the next 10 years as a function of the decrease in LDL-C achieved with statin therapy at various levels of absolute CVD risk. The absolute CVD risk threshold for the introduction of statin therapy in primary prevention varies throughout the world [8–11] (see later), but as long as the absolute CVD risk is known or can be estimated [9–11,13 ,15,16] and the pretreatment LDL-C is known, the NNT for the introduction of statin treatment can be calculated by applying the same equation (Table 1). &
NUMBER NEEDED TO TREAT AND CURRENTLY RECOMMENDED PRETREATMENT LDL CHOLESTEROL CONCENTRATIONS AND GOALS OF STATIN THERAPY Once the decision has been made to introduce a statin, the aim of treatment in primary prevention is
(b) 182
182
100
100
80
60 10 40 15 20 20
Cost-effective
5
NNT to prevent one event in next 10 years
NNT to prevent one event in next 10 years
5
30
0
80
60 10 40 15 20 20
30
0 0.5
1.0
1.5
2.0
2.5
3.0
LDL cholesterol decrease (mmol/l)
0.5
1.0
1.5
2.0
2.5
3.0
LDL cholesterol decrease (mmol/l)
FIGURE 1. (a) The number of people who must be treated to prevent one event (NNT) in the next 10 years as a function of the LDL-C decrease achieved with statin treatment at 5, 10, 15, 20 and 30% absolute atherosclerotic cardiovascular disease (CVD) risk in the same 10-year period. (b) The NNT in the next 10 years as a function of the LDL-C decrease achieved with statin treatment at different 10-year CVD risks as in Figure 1a but in this figure people whose NNT is 19 or less whose 10-year CVD risk is less than 20% are shaded. The NNT of 19 would be achieved in someone whose 10-year risk was 20% (currently recommended for statin treatment) whose LDL-C was lowered by 1.2 mmol/l using simvastatin 40 mg daily [see text for explanation]. 0957-9672 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-lipidology.com
241
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Hyperlipidaemia and cardiovascular disease Table 1. The number needed to treat (NNT) to prevent one CVD event (fatal or nonfatal) with statin therapy according to various primary prevention recommendations at pretreatment LDL-C levels of 3, 4 and 5 mmol/l with the goal of achieving LDL-C 2.5 mmol/l. CVD morbidity is assumed to be two to three times greater than mortality in calculating NNT for ESC/EAS recommendations (see discussion)
End point predicted
NNT to prevent one event Pretreatment LDL cholesterol
Critical 10-year absolute risk for introduction of statin
3 mmol/l
4 mmol/l
5 mmol/l
CVD mortality and morbidity NICE, JBS2
20%
45
15
9
Proposed NICE
10%
91
30
18
ACC/AHA
7.5%
121
40
24
5%
45–61
15–20
9–12
CVD mortality ESC/EAS
&
ACC/AHA, American College of Cardiology/American Heart Association [13 ]; ESC/EAS, European Society of Cardiology/European Atherosclerosis Society [10]; JBS2, Second Joint British Societies [11]; NICE, National Institute for Health and Clinical Excellence [8]; Proposed NICE [14].
to achieve LDL-C concentrations of less than 2.5 mmol/l (
The importance of considering LDL cholesterol response as well as cardiovascular risk in deciding who can benefit from statin therapy Handrean Soran a,b, Jonathan D. Schofield a,b, and Paul N. Durrington a
Purpose of review Guidelines seeking to deploy statin treatment rely heavily on the use of estimates of absolute cardiovascular disease (CVD) risk as an arbiter of who should receive statins. We question whether this is an effective strategy unless the LDL-cholesterol (LDL-C) response is also considered. Recent findings Recently, meta-analyses of randomized clinical trials of statins have revealed that CVD risk decreases linearly by 22% for each 1 mmol/l reduction in LDL-C. Calculation of the number needed to treat with statins to prevent one CVD event using both the pretreatment absolute CVD risk and the LDL-C response that can be achieved is thus possible. Application of this evidence reveals that many people (including younger ones) with high LDL-C levels can benefit more than people currently receiving statin treatment solely on the basis of their absolute CVD risk, whereas others at higher CVD risk, but with lower LDL-C, will derive little benefit. This does not seem to have been adequately considered in recent clinical guidelines. Summary A simple additional mathematical step in risk assessment to take account of the LDL-C response to statins and provide knowledge of number needed to treat would greatly improve individual management, costeffectiveness and the population impact of statins. Keywords absolute CVD risk, LDL cholesterol, number needed to treat, population impact of statin therapy
INTRODUCTION In a meta-analysis of 26 randomized clinical trials involving 170 000 participants, statins have been shown to decrease atherosclerotic cardiovascular disease (CVD) morbidity and mortality [1]. In this meta-analysis, the definition of CVD included coronary heart disease and cerebrovascular clinical events (both fatal and nonfatal), the primary end point of most of the trials. The typical reduction in LDL-cholesterol (LDL-C) in trials comparing active statin treatment with placebo was 1.07 mmol/l. In many of these trials, less potent statins were used than were later developed. Even so, the overall decrease in the rate of new CVD events (coronary heart disease and stroke) was 24%, which was highly statistically significant [1,2]. This 24% risk reduction is a relative risk reduction in the sense that it compares the event rate in the statin-treated (actively treated) trial participants with the event rate in those receiving placebo. The absolute number of events prevented during a period of treatment with statins (in other words the actual benefit) will
depend on the event rate in the controls receiving placebo (the rate that would have occurred had statin not been administered). This figure is frequently expressed in guidelines as the number of CVD events occurring in 100 people over the next 10 years, often termed the percentage (%) 10-year risk. Thus, statin treatment in a group of people whose absolute CVD risk is 20% (i.e. 20 per 100) over the next 10 years would be expected to decrease the anticipated rate by 20 0.24%, thus preventing
a Cardiovascular Research Group, School of Biomedicine, University of Manchester, and bCardiovascular Trials Unit, University Department of Medicine, Central Manchester University Hospitals NHS Foundation Trust, Manchester, UK
Correspondence to Professor Paul N. Durrington Cardiovascular Research Group, School of Biomedicine, Core Technology Facility (3rd Floor), University of Manchester, 46 Grafton Street, ManchesterM13 9NT, UK. Tel: +44 161 275 1201; fax: +44 161 275 1183; e-mail: [email protected] Curr Opin Lipidol 2014, 25:239–246 DOI:10.1097/MOL.0000000000000097
0957-9672 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-lipidology.com
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Hyperlipidaemia and cardiovascular disease
KEY POINTS Many people whose absolute CVD risk is considered too low for statin therapy would benefit more than those at higher risk, but with lower LDL-C. Number needed to treat (NNT) is a better way of assessing who should receive statin treatment than simply absolute CVD risk. CVD risk decreases linearly by 22% for each 1 mmol/l reduction in LDL-C achieved with statin treatment. NNT for 10 years to prevent one CVD event ¼ 100 (absolute CVD risk LDL-C decrease 0.22), in which absolute CVD risk is the percentage over the next 10 years and LDL-C decreases in millimoles per litre (multiply NNT by 39, if milligrams per decilitre is the unit used for cholesterol).
4.8 events per 100 people treated. The number of people who must receive treatment for 10 years to prevent one event is calculated by dividing 100 by 4.8, which, in this example, approximates to 21. This figure is referred to as the NNT. When the quality of life of those spared with a CVD event is unimpaired, as it is in the case of primary prevention of CVD with a statin, the NNT and, of course, the cost of treatment itself are the major determinants of the cost of a quality-adjusted life year (QALY) [3]. The relative CVD risk reduction with statin treatment (expressed as the ratio between the rate in treated participants to that in the controls receiving placebo or less efficacious statin treatment in terms of LDL-C lowering) is not constant and varies with the degree to which LDL-C concentration is decreased [1]. The statin trials overall reveal that relative CVD risk declines linearly with respect to the therapeutically induced reduction in LDL-C [1,4]. Meta-analysis further reveals that the percentage decrease in CVD is 22% (95% CI 20, 24) for each 1 mmol/l reduction in LDL-C [1]. The decreased CVD incidence with statins relates most strongly to effects on LDL-C [5] and, although statins differ as to their action on HDL-C levels, this is inconsequential [6 ]. The linear decrease of 22% in relative CVD incidence for each 1 mmol/l reduction in LDL-C does seem to persist down to statin-induced levels of 2 mmol/l and possibly lower [1]. Furthermore, it also applies regardless of the major source of CVD risk, be it diabetes, high blood pressure, smoking, low HDL-C, raised triglycerides or previous CVD. Statins, such as atorvastatin and rosuvastatin, can achieve decreases in LDL-C of more than 50% [7]. International and national clinical recommendations for statin treatment have frequently not taken into account the degree of LDL-C reduction that can be
achieved and have focussed very much on absolute CVD risk as an indication for statin therapy. Misleading conclusions have been reached, particularly in primary prevention, which have major consequences for optimal cost-effective statin deployment in the population and the treatment of individuals.
CURRENT AND PROPOSED RECOMMENDATIONS FOR STATIN TREATMENT The value of statin treatment to reduce CVD is not disputed, but current national and international guidelines focus on baseline CVD risk as the main criterion for prescribing statins [8–11]. Because most CVD events occur in people at average CVD risk simply because they are the most numerous (Rose effect) [12], it has been proposed both by the American College of Cardiology and American Heart Association (ACC/AHA) and by the National Institute for Health and Care Excellence (NICE) that statin use should no longer be confined to particularly high-risk individuals, but extended to people whose risk, at least from middle age onwards, is little more than the population average [13 ,14]. The evidence has still largely been neglected that the benefits of statin therapy relate to the extent of LDL-C reduction, which is greater in those with higher LDL-C and with more potent statins. This review considers a method of taking into account the extent of LDL-C lowering which can be achieved in individual patients, which, combined with the absolute CVD risk, permits the calculation of NNT and provides a more discriminating means of determining who should receive statin treatment. We are led to conclude that a simple formula to modify current estimates of absolute CVD risk estimation to provide NNT would greatly improve clinical guidance, particularly in primary prevention, by more accurately identifying people who can derive substantial benefit, including some from sections of the population at lower absolute risk, in whom more events can be prevented, than in many higher risk patients with lower LDL-C levels. &
&&
240
www.co-lipidology.com
CALCULATION OF NUMBER NEEDED TO TREAT The formula for calculating NNT to prevent one CVD event with statin therapy, taking into account the decrease in LDL-C achieved, is derived as follows: Events prevented per 100 people (e) ¼ Absolute CVD risk LDL-C decrease 0.22 NNT ¼ 100 e NNT ¼ 100 ðabsolute CVD risk LDL cholesterol decrease 0:22Þ % over next 10 years: mmol=l: Volume 25 Number 4 August 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
LDL response as a statin indication Soran et al.
APPLICATION OF NUMBER NEEDED TO TREAT TO STATIN INDICATIONS To demonstrate the use of the equation to calculate the statin NNT in primary prevention, taking into account the extent of LDL-C lowering, let us calculate the NNT for someone with no history of CVD who has an absolute CVD risk of 20% over the next 10 years. If the pretreatment LDL-C is, say, 3.5 mmol/l, decreasing this by 1 mmol/l to 2.5 mmol/l will prevent 4.4 CVD events per 100 similar people treated (e ¼ 20 1 0.22), meaning the number needed to receive treatment to prevent one event is 23, that is, the NNT is 23 (100 4.4). Suppose we now use this formula to calculate the NNT for someone at a 10 year CVD risk of 10% whose pretreatment LDL-C is 4.5 mmol/l. Achieving an LDL-C goal of 2.5 mmol/l interestingly also results in an NNT of 23 (100 (10 2 0.22) ¼ 23). This patient, however, would not, for example, have received statin treatment according to current NICE guidance, which states that patients must have a 10-year CVD risk of 20% or more to be treated [8]. If the NNT is calculated for someone with a pretreatment LDL-C of 5.5 mmol/l, reaching a target LDL-C of 2.5 mmol/l has an NNT value of 22, similar to the
(a)
earlier two examples, even when the 10-year CVD risk is as low as 7%. The degree to which LDL-C can be lowered should therefore be taken into account, as well as the CVD risk, in deciding who should receive statin therapy. Figure 1a shows the NNT to prevent one event in the next 10 years as a function of the decrease in LDL-C achieved with statin therapy at various levels of absolute CVD risk. The absolute CVD risk threshold for the introduction of statin therapy in primary prevention varies throughout the world [8–11] (see later), but as long as the absolute CVD risk is known or can be estimated [9–11,13 ,15,16] and the pretreatment LDL-C is known, the NNT for the introduction of statin treatment can be calculated by applying the same equation (Table 1). &
NUMBER NEEDED TO TREAT AND CURRENTLY RECOMMENDED PRETREATMENT LDL CHOLESTEROL CONCENTRATIONS AND GOALS OF STATIN THERAPY Once the decision has been made to introduce a statin, the aim of treatment in primary prevention is
(b) 182
182
100
100
80
60 10 40 15 20 20
Cost-effective
5
NNT to prevent one event in next 10 years
NNT to prevent one event in next 10 years
5
30
0
80
60 10 40 15 20 20
30
0 0.5
1.0
1.5
2.0
2.5
3.0
LDL cholesterol decrease (mmol/l)
0.5
1.0
1.5
2.0
2.5
3.0
LDL cholesterol decrease (mmol/l)
FIGURE 1. (a) The number of people who must be treated to prevent one event (NNT) in the next 10 years as a function of the LDL-C decrease achieved with statin treatment at 5, 10, 15, 20 and 30% absolute atherosclerotic cardiovascular disease (CVD) risk in the same 10-year period. (b) The NNT in the next 10 years as a function of the LDL-C decrease achieved with statin treatment at different 10-year CVD risks as in Figure 1a but in this figure people whose NNT is 19 or less whose 10-year CVD risk is less than 20% are shaded. The NNT of 19 would be achieved in someone whose 10-year risk was 20% (currently recommended for statin treatment) whose LDL-C was lowered by 1.2 mmol/l using simvastatin 40 mg daily [see text for explanation]. 0957-9672 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-lipidology.com
241
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Hyperlipidaemia and cardiovascular disease Table 1. The number needed to treat (NNT) to prevent one CVD event (fatal or nonfatal) with statin therapy according to various primary prevention recommendations at pretreatment LDL-C levels of 3, 4 and 5 mmol/l with the goal of achieving LDL-C 2.5 mmol/l. CVD morbidity is assumed to be two to three times greater than mortality in calculating NNT for ESC/EAS recommendations (see discussion)
End point predicted
NNT to prevent one event Pretreatment LDL cholesterol
Critical 10-year absolute risk for introduction of statin
3 mmol/l
4 mmol/l
5 mmol/l
CVD mortality and morbidity NICE, JBS2
20%
45
15
9
Proposed NICE
10%
91
30
18
ACC/AHA
7.5%
121
40
24
5%
45–61
15–20
9–12
CVD mortality ESC/EAS
&
ACC/AHA, American College of Cardiology/American Heart Association [13 ]; ESC/EAS, European Society of Cardiology/European Atherosclerosis Society [10]; JBS2, Second Joint British Societies [11]; NICE, National Institute for Health and Clinical Excellence [8]; Proposed NICE [14].
to achieve LDL-C concentrations of less than 2.5 mmol/l (
