This cohort study examines data from the Copenhagen General Population Study to compare the clinical performance of the 2021 European Society of Cardiology guidelines about statin use with guidelines from the American College of Cardiology/American Heart Association, UK National Institute for Health and Care Excellence, and previous European guidelines.
Key Points
Question
What is the clinical performance of the 2021 European Society of Cardiology (ESC) guidelines on using statins in primary prevention of atherosclerotic cardiovascular disease compared with other international guidelines?
Findings
In this cohort study representing 66 909 apparently healthy individuals, 4% qualified for statins according to the 2021 ESC guidelines compared with 20% to 34% according to guidelines from the ESC/European Atherosclerosis Society in 2019, American College of Cardiology/American Heart Association, and UK National Institute for Health and Care Excellence. The clinical performance of the 2021 ESC guidelines could be improved by lowering the treatment threshold for statins.
Meaning
The 2021 ESC guidelines dramatically reduce statin eligibility in low-risk European countries, and future European guidelines should consider lower treatment thresholds to increase the use of statins for primary prevention of cardiovascular disease.
Abstract
Importance
For primary prevention of atherosclerotic cardiovascular disease (ASCVD), the 2021 European Society of Cardiology (ESC) guidelines on statin use (hereafter European-ESC) recommend a new risk model (Systematic Coronary Risk Evaluation 2 [European-SCORE2]) as well as new age-specific treatment thresholds (≥7.5% 10-year ASCVD risk if aged 40-49 years and ≥10% if aged 50-69 years).
Objective
To compare the clinical performance of the 2021 European-ESC, American College of Cardiology/American Heart Association (hereafter US-ACC/AHA), UK National Institute for Health and Care Excellence (UK-NICE), and 2019 ESC/European Atherosclerosis Society (EAS) guidelines in apparently healthy individuals.
Design, Setting, and Participants
This population-based contemporary cohort study included 66 909 individuals from the Copenhagen General Population Study. Participants were aged 40 to 69 years and were free of ASCVD, diabetes, chronic kidney disease, and statin use at baseline in 2003 to 2015. Mean follow-up time was 9.2 years. Data were analyzed from November 2021 to April 2022.
Exposures
Statin treatment according to guideline criteria.
Main Outcomes and Measures
Calibration of risk calculators, statin eligibility, sensitivity, and specificity for ASCVD events according to guideline criteria.
Results
During follow-up, a range of 2962 to 4277 nonfatal and fatal ASCVD events was observed, as defined by the 2021 European-SCORE2, US pooled cohort equations (PCE), and UK-QRISK3 models, and 180 fatal ASCVD events were noted as defined by the 2019 European-SCORE1 model. European-SCORE2 was slightly better calibrated with a predicted/observed ASCVD event ratio of 0.8 vs 1.3 for US-PCE, 1.3 for UK-QRISK3, and 5.8 for European-SCORE1. For primary prevention class I recommendations in individuals aged 40 to 69 years, 2862 of 66 909 (4%) qualified for statins according to the 2021 European-ESC guidelines compared with 23 029 (34%) with US-ACC/AHA, 17 659 (26%) with UK-NICE, and 13 496 (20%) with 2019 European-ESC/EAS guidelines, with associated sensitivities for detecting future European-SCORE2–defined ASCVD events of 12%, 60%, 51%, and 36%, respectively. The sensitivity of the European-ESC guidelines was improved considerably by lowering the treatment thresholds, resulting in smaller losses in specificity. To obtain similar clinical performance with the 2021 European-ESC guidelines as in the other guidelines, the threshold with European-SCORE2 should be reduced to 5% overall to match US-ACC/AHA, to 6% to match UK-NICE, and to 7% to match 2019 European-ESC/EAS guidelines.
Conclusions and Relevance
Despite an improved European-SCORE2 prediction model, the new treatment thresholds in the 2021 European-ESC guidelines dramatically reduce eligibility for primary prevention with statins in low-risk European countries. Using lower treatment thresholds can improve overall guideline performance.
Introduction
The European Society of Cardiology (ESC) guidelines and other international guidelines on primary prevention of atherosclerotic cardiovascular disease (ASCVD) follow the principle of matching the intensity of preventive treatment to the absolute risk of patients.1 Thus, treatment with statins is restricted to those who have 10-year risk for ASCVD above certain guideline-defined treatment thresholds.2
Accordingly, the cornerstone of this risk-based strategy is the risk model used for risk assessment. In Europe, the first Systematic Coronary Risk Evaluation model (hereafter European-SCORE1) has been used since the 2003 ESC guidelines.3 European-SCORE1 estimates the 10-year risk for fatal ASCVD. Using only fatal ASCVD has several limitations, including underestimation of total ASCVD risk.4 In contrast, the models used by the American College of Cardiology/American Heart Association (hereafter US-ACC/AHA) and UK National Institute for Health and Care Excellence (UK-NICE) guidelines, that is, the pooled cohort equations (US-PCE) and UK-QRISK models, estimate the 10-year risk of total ASCVD events (nonfatal and fatal ASCVD events).5,6 Therefore, the US-ACC/AHA and UK-NICE guidelines have been able to define treatment thresholds that are based on both evidence from randomized clinical trials of statins as well as risk-benefit considerations with improved clinical performance compared with the SCORE1-based European-ESC guidelines.1,7,8 Recently, the European-SCORE2 model, predicting fatal and nonfatal ASCVD events,9 was developed and incorporated in the 2021 European-ESC prevention guidelines together with new age-specific treatment thresholds for statins.9
We compared the 2021 European-ESC prevention guidelines with the previous 2019 European-ESC/European Atherosclerosis Society (EAS) dyslipidemia guidelines as well as with the current US-ACC/AHA and UK-NICE guidelines. Because the 2021 European-ESC guidelines only provide class I recommendation for statins to individuals aged 40 to 69 years, we restricted all analyses to this age range to make fair comparisons of guidelines in their intended age range.
Methods
The Copenhagen General Population Study (CGPS) is a prospective cohort study of the Danish general population.10 Participants were randomly selected through the Danish Civil Registration System to reflect the Danish general population. For this study, we included 66 909 apparently healthy White individuals aged 40 to 69 years who were enrolled from 2003 through 2015. We excluded individuals with preexisting ASCVD, diabetes, chronic kidney disease (estimated glomerular filtration rate <60 mL/min/1.73 m2), or statin use and those with missing covariate data at the baseline examination. Additional details appear in the eMethods in the Supplement. The study was approved by Herlev and Gentofte Hospital and by a Danish ethics committee. Written informed consent was obtained from all individuals.
Risk Algorithms and Predicted Outcomes
The European-SCORE2 model predicts total ASCVD, that is, fatal ASCVD (similar end point as in the SCORE1 model plus nonfatal myocardial infarction and stroke).9 For the present study, we used the low-risk version of European-SCORE2 intended for most of western Europe, including Denmark, Norway, the Netherlands, Belgium, the United Kingdom, France, Spain, Switzerland, Luxembourg, and Israel. Countries are considered low risk by ESC guidelines if their World Health Organization age- and sex-standardized cardiovascular disease mortality rates are less than 100 cardiovascular disease deaths per 100 000 persons. The population of low-risk European countries is more than 230 million people.
The US-PCE predicts the 10-year risk jointly for nonfatal myocardial infarction, fatal coronary heart disease, and stroke.11,12 The UK-QRISK model recommended in the United Kingdom estimates the 10-year risk for an expanded ASCVD end point that in addition to nonfatal myocardial infarction, fatal coronary heart disease, and stroke also includes non–hard ASCVD end points (ie, angina) in the predicted outcome.6,13 QRISK3 is the most recent version of QRISK.14
The European-SCORE1 model predicts 10-year risk of fatal ASCVD.3 The fatal ASCVD end point includes death due to coronary heart disease, stroke, and other sudden death excluding definite nonatherosclerotic causes of death. For the present study, we used the 2019 low-risk version of European-SCORE1.15
Guideline-Defined Treatment Thresholds for Statin Therapy
In the present study, we compared class I and strong recommendations for primary prevention with statins. In sensitivity analyses, we also included weaker class II recommendations.
The 2021 European-ESC guidelines recommend age-specific treatment thresholds. Class I recommendations are provided for patients aged 40 to 49 years with European-SCORE2 10-year ASCVD risk of 7.5% or greater and for patients aged 50 to 69 with European-SCORE2 risk 10% or greater. For patients 70 years and older, there are no class I recommendations. Class II recommendations are provided for patients aged 40 to 49 years with European-SCORE2 risk 2.5% or greater and for patients aged 50 to 69 with European-SCORE2 risk 5% or greater in the presence of risk modifiers (ie, ankle-brachial index <0.9, low socioeconomical status, waist >94 cm in women and >102 cm in men, family history of ASCVD). We were not able to consider other suggested risk modifiers, such as coronary artery calcium, carotid ultrasound, or polygenic risk score, because the data were not available in the CGPS.
Based on the US-PCE model, the 2013 + 2019 US-ACC/AHA guidelines use a 10-year ASCVD risk threshold of 7.5% or greater to identify individuals who are eligible to take statins (class I recommendation).11,12 Class II recommendations are additionally provided for those with US-PCE of 5% up to 7.5% if risk enhancers are present. We were able to assess the following risk enhancers: chronic kidney disease, rheumatoid arthritis, low-density lipoprotein (LDL) cholesterol values 158 mg/dL or higher, plasma triglycerides 176 mg/dL or higher, C-reactive protein 2.0 mg/dL or higher, lipoprotein(a) values higher than 50 mg/dL, and apolipoprotein B values of 130 mg/dL or higher. (To convert LDL cholesterol to millimoles per liter, multiply by 0.0259; triglycerides to millimoles per liter, multiply by 0.0113.) We were not able to consider other risk enhancers such as preeclampsia, premature menopause, other inflammatory diseases, or coronary artery calcium.
Based on the UK-QRISK3 model, the 2014 UK-NICE guidelines use a 10-year ASCVD risk of 10% or greater to identify patients eligible for statin therapy (strong recommendation).6,16 The UK-NICE guidelines do not provide class II recommendations.
Based on the European-SCORE1 model, the European-ESC/EAS guidelines from 2003 + 2012 + 2016 + 2019 on the primary prevention of ASCVD recommend a 10-year fatal ASCVD risk of 5% or greater for identifying patients eligible for statins.3,15,17,18,19 Among patients with European-SCORE1 risk below 5%, those with LDL cholesterol 190 mg/dL or higher had class I recommendations.15 For patients with European-SCORE1 risk of 5% to 10% and 10% or greater, the 2019 European-ESC/EAS guidelines provide class I recommendations for statins to those with LDL cholesterol 100 mg/dL or higher and 70 mg/dL or higher, respectively. Class II recommendations are provided for patients with European-SCORE1 risk 1% to 5% with LDL cholesterol 100 mg/dL or higher, to patients with European-SCORE1 risk 5% to 10% with LDL cholesterol 70 mg/dL or higher, and to patients with European-SCORE1 risk 10% or greater with LDL cholesterol 55 mg/dL or higher.
End Points
For assessing calibration of the risk models, we closely followed the end point definitions used by each model. Every individual in Denmark is assigned a personal identification number at birth or immigration by which they can be traced in the national registries. We can account for everyone at all time points until death, emigration, occurrence of ASCVD, or end of follow-up in December 2018, whichever occurred first.
Statistical Analyses
We used Stata version 15.1 SE. Baseline characteristics are presented as proportions for categorical variables and as medians (IQR) for continuous variables. Estimated 10-year risk of ASCVD events was calculated for all participants in the CGPS using the European-SCORE2, US-PCE, UK-QRISK3, and European-SCORE1 risk models.
Calibration of the risk models was assessed by the ratio of predicted-to-observed (P/O) ASCVD events as well as by Hosmer-Lemeshow χ2 goodness-of-fit. For the calibration analyses, we used the end points defined by the risk algorithms. As not all individuals in the CGPS had completed 10 years of follow-up, the calibration analyses were restricted to the 54 365 individuals with a full 10 years of follow-up.
Discrimination of the risk models was compared for the prediction of European-SCORE2–defined ASCVD events with Harrell C statistics for prospective analyses. Clinical performance of the different guideline recommendations was estimated as the proportion of patients qualifying for statin therapy. Further, we estimated the sensitivity and specificity of guideline recommendations for capturing the patients who would later develop a European-SCORE2–defined ASCVD event. Clinical performance at different treatment thresholds ranging from 1% to 20% 10-year ASCVD risk with the 2021 European-SCORE2 model was assessed as the proportion of individuals who would be statin eligible.
Results
Baseline characteristics of the 66 909 CGPS individuals aged 40 to 69 years are shown in eTable 1 in the Supplement. The age distribution of the CGPS is similar to the entire Danish population, and the age distribution of the Danish population is similar to that in the other low-risk countries (eFigure 1 and 2 in the Supplement). During a mean follow-up of 9.2 years, 2692 individuals experienced a European-SCORE2 event, 2782 a US-PCE event, and 4277 a UK-QRISK3 event, while 180 experienced a European-SCORE1 event.
Estimated 10-Year Risk
There was a high degree of correlation between estimated risk with the different models (Spearman coefficients of 0.88 to 0.99; eFigure 3 in the Supplement). Estimating risk with the US-PCE and UK-QRISK3 models resulted in higher risk than with the European-SCORE2 model (Table 1 and eFigure 3 in the Supplement). Although the European-SCORE2 model estimates total ASCVD risk, many individuals had lower estimated risk with European-SCORE2 than with the fatal ASCVD-based European-SCORE1 (eFigure 3 in the Supplement). As shown in eFigure 4 through 6 in the Supplement, higher age is a stronger driver of risk with European-SCORE1 than with European-SCORE2 in both sexes. Likewise, higher blood pressure increases European-SCORE1 risk more than European-SCORE2 risk in men, while the opposite is true for cholesterol levels in both sexes. Accordingly, the higher risk with European-SCORE1 relative to European-SCORE2 was observed primarily in individuals older than 58.5 years (eFigure 7 in the Supplement).
Table 1. Predictive Performance of the 2021 European-SCORE2, US-PCE, UK-QRISK3, and European-SCORE1 Models.
| Risk model | Mean estimated 10-y ASCVD risk, % (95% CI) | P/O ratio | HL coefficienta | Harrel C statisticb | P value |
|---|---|---|---|---|---|
| European-SCORE2 | 4.2 (4.1-4.2) | 0.8 | 207 | 0.710 [Reference] | |
| US-PCE | 6.4 (6.4-6.5) | 1.3 | 442 | 0.712 | .004 |
| UK-QRISK3 | 7.4 (7.4-7.5) | 1.3 | 460 | 0.713 | .005 |
| European-SCORE1 | 2.1 (2.0-2.1) | 5.8 | 957 | 0.700 | <.001 |
Abbreviations: ASCVD, atherosclerotic cardiovascular disease; HL, Hosmer-Lemeshow; PCE, pooled cohort equations; P/O, predicted/observed ASCVD events; SCORE, Systematic Coronary Risk Evaluation.
For the Hosmer-Lemeshow goodness of fit, lower scores indicate better fit.
Discrimination (C statistic) for European-SCORE2 ASCVD end point.
Calibration and Discrimination
The overall P/O ratio was 0.8 for European-SCORE2 compared with 1.3 for US-PCE, 1.3 for UK-QRISK3, and 5.8 for European-SCORE1 (Table 1). The underestimation of risk for the European-SCORE2 model was most pronounced in individuals with high risk while it was well calibrated in the 7 lowest deciles of risk (eFigure 8 in the Supplement). Similarly, the US-PCE and UK-QRISK3 models were well calibrated in the lower risk ranges, including near their guideline-defined treatment thresholds. In contrast, the European-SCORE1 model overestimated risk throughout the risk spectrum. Overall, European-SCORE2 was the risk model that showed the best fit while having similar discriminatory performance as the other models (Table 1 and eFigure 9 in the Supplement).
Clinical Performance of Guidelines
Overall, class I recommended statin eligibility was found for 2862 of 66 909 individuals (4%) according to the 2021 European-ESC guidelines, 23 029 (34%) with the US-ACC/AHA guidelines, 17 659 (26%) with the UK-NICE guidelines, and 13 496 (20%) with the 2019 European-ESC guidelines (Figure 1). The corresponding sensitivities for detecting future European-SCORE2-defined ASCVD events were 14%, 60%, 51%, and 36%, respectively. Statin eligibility, sensitivity, and specificity varied substantially by age and sex (Figure 2 and eFigure 10-11 in the Supplement). Although statin eligibility increased with age by all guidelines, this was least pronounced for the 2021 European-ESC guidelines. Thus, across all age groups and in both men and women, statin eligibility and sensitivity for identifying those individuals who later develop ASCVD was lowest for the 2021 European-ESC guidelines. For example, in women, close to 0% (11) met the 7.5% European-SCORE2 risk threshold in those aged 40 to 49 years while 312 (1%) met the 10% risk threshold in those aged 50 to 69 years (Figure 2 and Table 2). In men, 215 (2%) met the 7.5% European-SCORE2 risk threshold in those aged 40 to 49 while 2327 (13%) of those aged 50 to 69 met the 10% risk threshold.
Figure 1. Statin Eligibility for Primary Prevention of Atherosclerotic Cardiovascular Disease (ASCVD) and Sensitivity and Specificity for Detecting ASCVD Events According to Guideline-Defined Class I/Strong Recommendations in Individuals Aged 40 to 69 Years.
Based on 66 909 individuals from the Copenhagen General Population Study who were free of ASCVD, diabetes, chronic kidney disease, and statin use at baseline. ACC/AHA indicates American College of Cardiology and American Heart Association; EAS, European Atherosclerosis Society; ESC, European Society of Cardiology; NICE, National Institute for Health and Care Excellence.
Figure 2. Statin Eligibility for Primary Prevention of Atherosclerotic Cardiovascular Disease (ASCVD) Stratified by Sex and 5-Year Age Groups According to Guideline-Defined Class I/Strong Recommendations in Individuals Aged 40 to 69 Years.
Based on 66 909 individuals from the Copenhagen General Population Study who were free of ASCVD, diabetes, chronic kidney disease, and statin use at baseline. ACC/AHA indicates American College of Cardiology and American Heart Association; EAS, European Atherosclerosis Society; ESC, European Society of Cardiology; NICE, National Institute for Health and Care Excellence.
Table 2. Proportion of Individuals Eligible for Statin Therapy and Its Clinical Performance With Varying Risk Thresholds Using the New European-SCORE2 Model and Class I Recommendation.
| 10-y SCORE2 threshold, % | Overall | Women | Men | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Statin eligibility, % | Sensitivity, % | Specificity, % | Statin eligibility, % | Sensitivity, % | Specificity, % | Statin eligibility, % | Sensitivity, % | Specificity, % | |
| All ages | |||||||||
| 20 | 0 | 0 | 100 | 0 | 0 | 100 | 0 | 0 | 100 |
| 15 | 1 | 1 | 99 | 0 | 0 | 100 | 1 | 3 | 99 |
| 10 | 5 | 13 | 93 | 1 | 3 | 99 | 10 | 20 | 91 |
| 7.50 | 12 | 31 | 88 | 4 | 10 | 96 | 24 | 46 | 77 |
| 5 | 31 | 57 | 69 | 17 | 34 | 83 | 48 | 73 | 53 |
| 4 | 42 | 69 | 58 | 27 | 47 | 74 | 61 | 83 | 40 |
| 3 | 56 | 81 | 45 | 39 | 64 | 61 | 75 | 92 | 25 |
| 2 | 72 | 90 | 28 | 57 | 78 | 44 | 91 | 97 | 10 |
| 1 | 90 | 97 | 10 | 83 | 94 | 18 | 99 | 99 | 1 |
| Age 40-49 y | |||||||||
| 20 | 0 | 0 | 100 | 0 | 0 | 100 | 0 | 0 | 100 |
| 15 | 0 | 0 | 100 | 0 | 0 | 100 | 0 | 0 | 100 |
| 10 | 0 | 1 | 100 | 0 | 0 | 100 | 1 | 3 | 99 |
| 7.50 | 1 | 5 | 99 | 0 | 1 | 100 | 2 | 10 | 97 |
| 5 | 5 | 15 | 95 | 1 | 5 | 99 | 12 | 24 | 89 |
| 4 | 10 | 27 | 90 | 2 | 11 | 98 | 21 | 42 | 79 |
| 3 | 20 | 43 | 80 | 5 | 22 | 95 | 41 | 63 | 60 |
| 2 | 40 | 64 | 60 | 15 | 38 | 84 | 74 | 89 | 27 |
| 1 | 72 | 87 | 28 | 57 | 76 | 43 | 98 | 99 | 20 |
| Age 50-69 y | |||||||||
| 20 | 0 | 0 | 99 | 0 | 0 | 100 | 0 | 0 | 100 |
| 15 | 0 | 2 | 99 | 0 | 0 | 100 | 2 | 3 | 99 |
| 10 | 7 | 16 | 93 | 1 | 3 | 98 | 13 | 24 | 86 |
| 7.50 | 20 | 36 | 81 | 7 | 12 | 93 | 35 | 52 | 66 |
| 5 | 46 | 66 | 56 | 27 | 41 | 72 | 68 | 82 | 34 |
| 4 | 60 | 77 | 41 | 42 | 57 | 57 | 82 | 91 | 19 |
| 3 | 75 | 88 | 26 | 60 | 76 | 39 | 93 | 97 | 8 |
| 2 | 89 | 95 | 12 | 80 | 90 | 19 | 98 | 99 | 2 |
| 1 | 99 | 99 | 2 | 98 | 99 | 2 | 99 | 99 | 1 |
Abbreviation: SCORE, Systematic Coronary Risk Evaluation.
eTable 2 in the Supplement shows baseline characteristics of individuals who were statin eligible with the 2019 and 2021 European-ESC guidelines, including those with discordant recommendations. Only 282 individuals (0.4%) became newly statin eligible with the 2021 ESC vs 2019 ESC/EAS guidelines (eTable 2 in the Supplement). These individuals were predominantly younger (median age, 49 years), smoking (83%), male (84%) individuals with high blood pressure and high estimated risk with US-PCE (18%). In contrast, 10 916 individuals (16% of the population) went from statin eligible in the 2019 ESC/EAS guidelines to statin noneligible in the 2021 ESC guidelines. The proportion of women in this group was higher at 36% and most were nonsmokers (16%) with lower estimated US-PCE risk (13%). Large differences in clinical performance between guidelines were also present when both class I/strong and class II recommendations were examined together (eFigure 12 in the Supplement).
Clinical Performance of Different European-SCORE2 Treatment Thresholds
Table 2 shows statin eligibility, sensitivity, and specificity for women and men, separately, based on varying European-SCORE2 treatment thresholds from 20% to 1% stratified by the age groups 40 to 49 and 50 to 69 years. Reducing the European-SCORE2 treatment threshold in the overall population (same threshold for all aged 40-69 years) or in an age- and sex-specific manner could substantially improve sensitivity for detecting future ASCVD events with smaller reductions in specificity. For example, reducing the risk threshold from 7.5% to 4% in men and to 2% in women aged 40 to 49 years markedly improved sensitivity (from 10% to 42% in men and from 1% to 38% in women) with only modest reduction in specificity (97% to 79% in men and 100% to 84% in women). Likewise, reduction of the risk threshold from 10% to 7.5% in men and to 5% in women improved sensitivity (from 24% to 52% in men and from 3% to 41% in women) with modest reductions in specificity.
To obtain similar clinical performance as other international guidelines, the European-SCORE2 threshold in individuals aged 40 to 69 years should be 5% to match US-ACC/AHA, 6% to match UK-NICE, and 7% to match 2019 European-ESC/EAS guidelines (Table 3 and eFigure 13-14 in the Supplement).
Table 3. Treatment Thresholds of 2021 European-SCORE2 Corresponding to Treatment Thresholds of US-PCE, UK-QRISK3, and European-SCORE1 in Individuals Aged 40 to 69 Years.
| Cut point (10-y risk) | Statin eligibility in the population, % | Performance for European-SCORE2–defined ASCVD eventsa | ||
|---|---|---|---|---|
| Sensitivity, % | Specificity, % | |||
| European-SCORE2 vs US-PCE, % | ||||
| SCORE2 | PCE | |||
| 10.0 | 20.7 | 5 | 12 | 95 |
| 9.8 | 20.0 | 6 | 14 | 93 |
| 5.1 | 7.5 | 32 | 57 | 69 |
| 3.9 | 5.0 | 44 | 71 | 56 |
| European-SCORE2 vs UK-QRISK3, % | ||||
| SCORE2 | QRISK3 | |||
| 10.0 | 20.2 | 5 | 12 | 95 |
| 5.8 | 10.0 | 26 | 50 | 75 |
| European-SCORE2 vs European-SCORE1, % | ||||
| SCORE2 | SCORE1 | |||
| 11.3 | 10.0 | 3 | 9 | 97 |
| 7.0 | 5.0 | 15 | 32 | 86 |
Abbreviations: ASCVD, atherosclerotic cardiovascular disease; PCE, pooled cohort equations; SCORE, Systematic Coronary Risk Evaluation.
European-SCORE2 end point of fatal ASCVD, nonfatal myocardial infarction, and stroke.
Discussion
In this large, contemporary, population-based European cohort of apparently healthy individuals, we compared the 2021 European-ESC guidelines on primary prevention using statins with the US-ACC/AHA, UK-NICE, and 2019 European-ESC/EAS guidelines. Although the 2021 European-ESC guidelines introduced a new and improved risk model, the updated age-specific recommendations dramatically reduce eligibility for class I recommended statins to only 4% with low sensitivity for detecting future ASCVD events. Notably, less than 1% of women met 2021 European-ESC treatment criteria. This is in sharp contrast to the previous 2019 European-ESC/EAS guidelines as well as current UK-NICE and US-ACC/AHA guidelines that provide class I/strong recommendations to approximately 20% to 34% of individuals with better clinical performance in both men and women. Importantly, we found that the clinical performance of the 2021 European-ESC guidelines could be improved by lowering the European-SCORE2 treatment threshold in either the overall population or by using age- and sex-specific thresholds. For obtaining similar clinical performance as in the US-ACC/AHA and UK-NICE guidelines, a European-SCORE2 risk threshold of 5% and 6% is needed, respectively. For obtaining status quo regarding the previous 5% European-SCORE1 threshold, a 7% threshold should be used with the European-SCORE2 model.
Our results are important for clinical practice and future European-ESC guidelines on primary prevention with statins as they point toward a likely unintended dramatic reduction in the potential for ASCVD prevention by implementing the 2021 European-ESC age-specific treatment criteria of 7.5% or greater in those aged 40 to 49 years and 10% or greater in those aged 50 to 69 years.
Improved Risk Prediction With 2021 European-SCORE2 vs 2019 European-SCORE1
The European-SCORE1 model differs from other risk models by only predicting risk of fatal ASCVD.3,5,13 Because treatment of established ASCVD is effective, most patients survive with chronic ASCVD.20,21 Thus, the European-SCORE1 model will substantially underestimate total ASCVD risk if it is well calibrated. However, total ASCVD risk estimated by the European-SCORE2 model is only slightly higher than risk estimated with European-SCORE1 model, and many individuals have a counterintuitively lower risk with the European-SCORE2 model than with the European-SCORE1 model. This paradoxical finding is explained by a combination of reasons: (1) that age is a stronger driver for risk in European-SCORE1 and (2) that the European-SCORE1 model hugely overestimates risk by almost 6-fold while the new European-SCORE2 model is better calibrated. The reason for this may be, at least partly, because the European-SCORE1 model (coefficients and survival function) was developed from historical cohorts undergoing baseline examination in 1967 to 1991. Since then, mortality from ASCVD has decreased substantially, leading to overestimation of fatal ASCVD risk with European-SCORE1.4 In contrast, the coefficients in the European-SCORE2 model were developed from more contemporary cohorts from 1990 to 2009 while the baseline survival function was recalibrated based on contemporary World Health Organization age- and sex-standardized rates.9
European-SCORE2 vs US-PCE and UK-QRISK
The European-SCORE2 and US-PCE models predict a similar ASCVD end point, although the European-SCORE2 end point also includes noncoronary cardiovascular deaths.5,9 Nevertheless, risk estimated with US-PCE was higher than with European-SCORE2. The calibration of the US-PCE has been questioned because some studies have indicated overestimation of risk22,23 while another study found it to be well calibrated.24 In our analyses, US-PCE overestimated risk among individuals with high estimated risk. More important for the intention of the US-ACC/AHA guidelines, however, the US-PCE model was well calibrated around the 7.5% risk threshold, indicating that it will correctly identify treatment candidates.
The European-SCORE2 model was developed to also be used in the United Kingdom as several UK cohorts were included in its development.9 Risk estimated with the UK-QRISK3 that includes non–hard ASCVD events yielded risks that are almost double those from European-SCORE2. As the UK-QRISK3 was relatively well calibrated, this demonstrates the importance of also considering non–hard ASCVD events when evaluating risk. The UK-QRISK3 was well calibrated around the treatment threshold, meaning that it will correctly identify treatment candidates according to UK-NICE guidelines.
Clinical Performance of Treatment Thresholds in the 2021 European-ESC Guidelines
The 2021 European-ESC guidelines have adopted a new age-specific treatment approach for the initiation of primary prevention with statins. The rationale behind these new treatment thresholds is not explained in the guidelines. Our analyses show that they substantially reduce statin eligibility and, for practical purposes, phase out class I recommendations for primary prevention in women.
In contrast, the risk-benefit–derived treatment thresholds adopted by US-ACC/AHA and UK-NICE identify 34% and 26% of individuals, respectively, who are statin eligible with an associated sensitivity for detecting future ASCVD events greater than 50%. Therefore, these guidelines have the potential to initiate preventive treatment for the majority of those who are destined to develop ASCVD. Cost-effectiveness analyses support these thresholds.25,26
Another important observation is that the clinical performance of the 2021 European-ESC guidelines could be improved by lowering the treatment threshold. Statin eligibility based on the European-SCORE2 is very sensitive to small changes in the selected treatment threshold. Thus, relatively small reductions in the selected treatment threshold can substantially improve sensitivity for detecting future ASCVD events with only minor losses in specificity. Similar results have previously also been found for the US-ACC/AHA guidelines.27 Further, if ASCVD is to be prevented to a similar degree in women as in men, then even lower sex-specific thresholds should be considered in women. However, no treatment threshold is perfect because improved sensitivity comes at the expense of lower specificity. Further, the use of 10-year risk thresholds will always favor initiation of prevention in older individuals. Thus, to improve ASCVD prevention in younger individuals, other approaches are needed, such as lower 10-year risk thresholds, considering risk enhancers, and, importantly, lifetime risk estimation.
Limitations
A potential limitation of our study is that we studied only White Europeans, and extrapolation of our results to racial and ethnic minority individuals living in low-risk European populations should be done cautiously. As Denmark is a low-risk European country, together with most of western Europe, our results should not be extrapolated to countries recommended to use the moderate, high-risk, or very-high-risk European-SCORE2. Further, the CGPS started recruitment in 2003, which could potentially make it less representative for more contemporary populations.
A major strength of our study is that the results originate from a contemporary, population-based large European cohort with not a single person lost to follow-up. Also, ASCVD events were appropriately identified, which is essential for the assessment of calibration of risk models. Further, our results are most likely generalizable to most other low-risk countries.
Conclusions
The newly introduced European-SCORE2 risk model performs better than the predecessor European-SCORE1 by incorporating nonfatal events, being better calibrated, and having improved discriminatory performance for ASCVD events. However, the new age-specific risk thresholds that were chosen for statin treatment in the 2021 European-ESC guidelines dramatically reduce statin eligibility to only 4% in the target population and essentially eliminate statin indication in women. Future ESC guidelines should consider lower thresholds for treatment to improve the clinical performance of the guidelines and align them with the US-ACC/AHA and UK-NICE guidelines.
eMethods
eReferences
eTable 1. Baseline characteristics of individuals in the Copenhagen General Population Study aged 40-69 years.
eTable 2. Comparison of baseline characteristics of individuals aged 40-69 years according to statin eligibility status with the 2019 European EAS/EAS and 2021 European-ESC guidelines
eFigure 1. Distribution of age among those aged 40-69 years in the Copenhagen General Population Study versus the entire Danish population per year 2022
eFigure 2. Distribution of age in Denmark versus the other low-risk European countries
eFigure 3. Correlation between risk estimated with 2021 European-SCORE2 model versus US-PCE, UK-QRISK3, and European-SCORE1 models
eFigure 4. 10-year European-SCORE1 and European-SCORE2 risk by varying single risk factor level in men
eFigure 5. 10-year European-SCORE1 and European-SCORE2 risk by varying single risk factor level in women
eFigure 6. 10-year European-SCORE1 and European-SCORE2 risk by varying systolic blood pressure in men and women aged 65 years
eFigure 7. Individual participant 10-year ASCVD risk difference between 2021 European-SCORE2 and 2019 European-SCORE1 models
eFigure 8. Calibration plots of the 2021 European-SCORE2, US-PCE, UK-QRISK3, and 2019 European-SCORE1 models
eFigure 9. Discriminatory performance of the European-SCORE2, US-PCE, UK-QRISK3, and European-SCORE1 models for predicting ASCVD events
eFigure 10. Sensitivity for detecting future European-SCORE2-defined ASCVD events of guideline-defined ClassI/strong recommendations for statin therapy
eFigure 11. Specificity of guideline-defined ClassI/strong recommendations for statin therapy for European-SCORE2-defined atherosclerotic cardiovascular disease events
eFigure 12. Statin eligibility and sensitivity and specificity for later ASCVD events using both I/strong and II recommendations according to the 2021 European-ESC, US-ACC/HA, UK-NICE, and 2019 European-ESC/EAS guidelines
eFigure 13. Statin eligibility across different treatment thresholds with the 2021 European-SCORE2, US-PCE, UK-QRISK3, and European-SCORE1 models
eFigure 14. Statin sensitivity and specificity for later ASCVD events across different treatment thresholds with 2021 European-SCORE2, US-PCE, UK-QRISK3, and European-SCORE1
References
- 1.Mortensen MB, Nordestgaard BG. Comparison of five major guidelines for statin use in primary prevention in a contemporary general population. Ann Intern Med. 2018;168(2):85-92. doi: 10.7326/M17-0681 [DOI] [PubMed] [Google Scholar]
- 2.Lloyd-Jones DM. Cardiovascular risk prediction: basic concepts, current status, and future directions. Circulation. 2010;121(15):1768-1777. doi: 10.1161/circulationaha.109.849166 [DOI] [PubMed] [Google Scholar]
- 3.Conroy RM, Pyörälä K, Fitzgerald AP, et al. ; SCORE project group . Estimation of ten-year risk of fatal cardiovascular disease in Europe: the SCORE project. Eur Heart J. 2003;24(11):987-1003. doi: 10.1016/S0195-668X(03)00114-3 [DOI] [PubMed] [Google Scholar]
- 4.Mortensen MB, Falk E. Limitations of the SCORE-guided European guidelines on cardiovascular disease prevention. Eur Heart J. 2017;38(29):2259-2263. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Goff DC Jr, Lloyd-Jones DM, Bennett G, et al. ; American College of Cardiology/American Heart Association Task Force on Practice Guidelines . 2013 ACC/AHA guideline on the assessment of cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2014;63(25 pt B):2935-2959. doi: 10.1016/j.jacc.2013.11.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.National Institute for Health and Care Excellence (NICE) National Clinical Guideline Centre . Clinical Guideline CG181: lipid modification: cardiovascular risk assessment and the modification of blood lipids for the primary and secondary prevention of cardiovascular disease. Updated November 21, 2016. http://www.nice.org.uk/Guidance/cg [PubMed]
- 7.Mortensen MB, Nordestgaard BG, Afzal S, Falk E. ACC/AHA guidelines superior to ESC/EAS guidelines for primary prevention with statins in non-diabetic Europeans: the Copenhagen General Population Study. Eur Heart J. 2017;38(8):586-594. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Mortensen MB, Nordestgaard BG. Statin use in primary prevention of atherosclerotic cardiovascular disease according to 5 major guidelines for sensitivity, specificity, and number needed to treat. JAMA Cardiol. 2019;4(11):1131-1138. doi: 10.1001/jamacardio.2019.3665 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.SCORE2 working group and ESC Cardiovascular risk collaboration . SCORE2 risk prediction algorithms: new models to estimate 10-year risk of cardiovascular disease in Europe. Eur Heart J. 2021;42:2439-2454. doi: 10.1093/eurheartj/ehab309 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Mortensen MB, Nordestgaard BG. Elevated LDL cholesterol and increased risk of myocardial infarction and atherosclerotic cardiovascular disease in individuals aged 70-100 years: a contemporary primary prevention cohort. Lancet. 2020;396(10263):1644-1652. doi: 10.1016/S0140-6736(20)32233-9 [DOI] [PubMed] [Google Scholar]
- 11.Stone NJ, Robinson JG, Lichtenstein AH, et al. ; American College of Cardiology/American Heart Association Task Force on Practice Guidelines . 2013 ACC/AHA guideline on the treatment of blood cholesterol to reduce atherosclerotic cardiovascular risk in adults: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2014;63(25 pt B):2889-2934. doi: 10.1016/j.jacc.2013.11.002 [DOI] [PubMed] [Google Scholar]
- 12.Grundy SM, Stone NJ, Bailey AL, et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol: Executive Summary: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2019;73(24):3168-3209. doi: 10.1016/j.jacc.2018.11.002 [DOI] [PubMed] [Google Scholar]
- 13.Hippisley-Cox J, Coupland C, Vinogradova Y, Robson J, May M, Brindle P. Derivation and validation of QRISK, a new cardiovascular disease risk score for the United Kingdom: prospective open cohort study. BMJ. 2007;335(7611):136-136. doi: 10.1136/bmj.39261.471806.55 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hippisley-Cox J, Coupland C, Brindle P. Development and validation of QRISK3 risk prediction algorithms to estimate future risk of cardiovascular disease: prospective cohort study. BMJ. 2017;357:j2099. doi: 10.1136/bmj.j2099 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Mach F, Baigent C, Catapano AL, et al. ; ESC Scientific Document Group . 2019 ESC/EAS guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk. Eur Heart J. 2019;37:2999. doi: 10.1016/j.atherosclerosis.2019.08.014 [DOI] [PubMed] [Google Scholar]
- 16.JBS3 Board . Joint British Societies’ consensus recommendations for the prevention of cardiovascular disease (JBS3). Heart. 2014;100(suppl 2):ii1-ii67. doi: 10.1136/heartjnl-2014-305693 [DOI] [PubMed] [Google Scholar]
- 17.De Backer G, Ambrosioni E, Borch-Johnsen K, et al. ; Third Joint Task Force of European and Other Societies on Cardiovascular Disease Prevention in Clinical Practice . European guidelines on cardiovascular disease prevention in clinical practice. Eur Heart J. 2003;24(17):1601-1610. doi: 10.1016/S0195-668X(03)00347-6 [DOI] [PubMed] [Google Scholar]
- 18.Perk J, De Backer G, Gohlke H, et al. ; European Association for Cardiovascular Prevention & Rehabilitation (EACPR); ESC Committee for Practice Guidelines (CPG) . European guidelines on cardiovascular disease prevention in clinical practice (version 2012): The Fifth Joint Task Force of the European Society of Cardiology and other societies on cardiovascular disease prevention in clinical practice (constituted by representatives of nine societies and by invited experts). Eur Heart J. 2012;33(13):1635-1701. [DOI] [PubMed] [Google Scholar]
- 19.Piepoli MF, Hoes AW, Agewall S, et al. ; ESC Scientific Document Group . 2016 European guidelines on cardiovascular disease prevention in clinical practice: the Sixth Joint Task Force of the European Society of Cardiology and other societies on cardiovascular disease prevention in clinical practice (constituted by representatives of 10 societies and by invited experts): developed with the special contribution of the European Association for Cardiovascular Prevention & Rehabilitation (EACPR). Eur Heart J. 2016;37(29):2315-2381. doi: 10.1093/eurheartj/ehw106 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Jørstad HT, Colkesen EB, Boekholdt SM, et al. Estimated 10-year cardiovascular mortality seriously underestimates overall cardiovascular risk. Heart. 2016;102(1):63-68. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Townsend N, Wilson L, Bhatnagar P, Wickramasinghe K, Rayner M, Nichols M. Cardiovascular disease in Europe: epidemiological update 2016. Eur Heart J. 2016;37(42):3232-3245. doi: 10.1093/eurheartj/ehw334 [DOI] [PubMed] [Google Scholar]
- 22.Kavousi M, Leening MJG, Nanchen D, et al. Comparison of application of the ACC/AHA guidelines, Adult Treatment Panel III guidelines, and European Society of Cardiology guidelines for cardiovascular disease prevention in a European cohort. JAMA. 2014;311(14):1416-1423. doi: 10.1001/jama.2014.2632 [DOI] [PubMed] [Google Scholar]
- 23.Cook NR, Ridker PM. Calibration of the pooled cohort equations for atherosclerotic cardiovascular disease: an update. Ann Intern Med. 2016;165:786-794. doi: 10.7326/M16-1739 [DOI] [PubMed] [Google Scholar]
- 24.Muntner P, Colantonio LD, Cushman M, et al. Validation of the atherosclerotic cardiovascular disease pooled cohort risk equations. JAMA. 2014;311(14):1406-1415. doi: 10.1001/jama.2014.2630 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Deaño RC, Pandya A, Jones EC, Borden WB. A look at statin cost-effectiveness in view of the 2013 ACC/AHA cholesterol management guidelines. Curr Atheroscler Rep. 2014;16(9):438. doi: 10.1007/s11883-014-0438-9 [DOI] [PubMed] [Google Scholar]
- 26.Pandya A, Sy S, Cho S, Weinstein MC, Gaziano TA. Cost-effectiveness of 10-year risk thresholds for initiation of statin therapy for primary prevention of cardiovascular disease. JAMA. 2015;314(2):142-150. doi: 10.1001/jama.2015.6822 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Navar-Boggan AM, Peterson ED, D’Agostino RB Sr, Pencina MJ, Sniderman AD. Using age- and sex-specific risk thresholds to guide statin therapy: one size may not fit all. J Am Coll Cardiol. 2015;65(16):1633-1639. doi: 10.1016/j.jacc.2015.02.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eMethods
eReferences
eTable 1. Baseline characteristics of individuals in the Copenhagen General Population Study aged 40-69 years.
eTable 2. Comparison of baseline characteristics of individuals aged 40-69 years according to statin eligibility status with the 2019 European EAS/EAS and 2021 European-ESC guidelines
eFigure 1. Distribution of age among those aged 40-69 years in the Copenhagen General Population Study versus the entire Danish population per year 2022
eFigure 2. Distribution of age in Denmark versus the other low-risk European countries
eFigure 3. Correlation between risk estimated with 2021 European-SCORE2 model versus US-PCE, UK-QRISK3, and European-SCORE1 models
eFigure 4. 10-year European-SCORE1 and European-SCORE2 risk by varying single risk factor level in men
eFigure 5. 10-year European-SCORE1 and European-SCORE2 risk by varying single risk factor level in women
eFigure 6. 10-year European-SCORE1 and European-SCORE2 risk by varying systolic blood pressure in men and women aged 65 years
eFigure 7. Individual participant 10-year ASCVD risk difference between 2021 European-SCORE2 and 2019 European-SCORE1 models
eFigure 8. Calibration plots of the 2021 European-SCORE2, US-PCE, UK-QRISK3, and 2019 European-SCORE1 models
eFigure 9. Discriminatory performance of the European-SCORE2, US-PCE, UK-QRISK3, and European-SCORE1 models for predicting ASCVD events
eFigure 10. Sensitivity for detecting future European-SCORE2-defined ASCVD events of guideline-defined ClassI/strong recommendations for statin therapy
eFigure 11. Specificity of guideline-defined ClassI/strong recommendations for statin therapy for European-SCORE2-defined atherosclerotic cardiovascular disease events
eFigure 12. Statin eligibility and sensitivity and specificity for later ASCVD events using both I/strong and II recommendations according to the 2021 European-ESC, US-ACC/HA, UK-NICE, and 2019 European-ESC/EAS guidelines
eFigure 13. Statin eligibility across different treatment thresholds with the 2021 European-SCORE2, US-PCE, UK-QRISK3, and European-SCORE1 models
eFigure 14. Statin sensitivity and specificity for later ASCVD events across different treatment thresholds with 2021 European-SCORE2, US-PCE, UK-QRISK3, and European-SCORE1