The PROCAM Study

Evaluating CMR - Assessing CVD risk: Traditional Approaches

Key Points

  • The PROCAM study showed that HDL cholesterol is an important modulator of CHD risk and provided evidence that HDL cholesterol levels exert a strong influence on CHD risk among individuals at elevated global cardiovascular risk.
  • An important contribution of the PROCAM study was to show that CHD risk factors do not act in isolation but rather in conjunction.
  • The PROCAM study demonstrated that fasting triglyceride levels are an independent risk factor for CHD events, regardless of HDL or LDL cholesterol concentrations.

The PROCAM Study

Over the last four decades, significant progress has been made in the identification of cardiovascular disease (CVD) risk factors. Identifying subjects at high risk of CVD is a key first step for CVD prevention [1]. Data from large prospective studies [2-5] have facilitated the identification of major CVD risk factors such as age, sex, smoking, high blood pressure, elevated total and LDL cholesterol levels, and low HDL cholesterol concentrations. The fact that risk factors work in conjunction to elevate CVD risk has been recognized for decades, but high-risk individuals have largely been identified based on single risk factors. Many groups and organizations have worked to develop approaches to identify high-risk individuals using global risk algorithms. The Framingham Heart Study [6], a landmark American prospective cardiovascular study, has provided a wealth of information on major coronary heart disease (CHD) risk factors and paved the way to the development of a simple CHD prediction algorithm by evaluating the presence and/or severity of several traditional risk factors. Another important study of cardiovascular epidemiology, the PROspective CArdiovascular Münster study (PROCAM), has also developed a scoring system for predicting global CHD risk in a large sample of subjects in Germany [7]. In contrast to the Framingham Heart Study, which was conducted in a relatively homogenous American population within a limited geographical area (the city of Framingham, Massachusetts), the PROCAM study developed a risk calculator tailored to a Northern European population.

Over the several waves of recruitment which started in 1978, the PROCAM study now involves about 50,000 participants aged 16 to 65 years from 84 companies and public service offices in the Münster and northern Ruhr areas of Germany [7]. Recruited volunteers had no history of myocardial infarction (MI) or stroke and no electrocardiogram (ECG) evidence of ischemic heart disease. Examination at baseline included a case history questionnaire, measurement of blood pressure and anthropometric data, a resting ECG, and determination of several laboratory parameters. Follow-up was performed by questionnaire, and cardiovascular events and deaths were confirmed by examining hospital records and/or eyewitness reports. The initial examination was repeated after 6 to 7 years.

As a whole, the PROCAM study was designed to study people for cardiovascular risk factors, mortality, and cardiovascular events (including MI and stroke). Its goal was to determine the prevalence of CVD risk factors in a German population and evaluate the relationship between risk factors and subsequent incidence of MI or death from CVD or other causes [8]. A key objective of the PROCAM study was therefore to improve the methods used to estimate CVD risk.

The PROCAM Risk Algorithm

The first PROCAM risk calculator was based on a scoring system for risk factors derived from 10-year follow-up data of the PROCAM study [7]. The algorithm was initially developed and validated in the cohort of 5,389 men (35 to 65 years of age) recruited before the end of 1985. Among those participants, 325 developed major coronary events within 10 years. The risk algorithm was expressed as a simple scoring scheme in order to make it relevant to clinical practice. To construct the algorithm, a Cox proportional hazards model was used. The prediction model includes age, LDL cholesterol, HDL cholesterol, triglycerides, smoking, diagnosis of diabetes, family history of MI, and systolic blood pressure. In contrast to the previous model generated by the Framingham Heart Study, the PROCAM score used triglycerides and family history of premature MI to calculate the risk score. Categorical variables are history of diabetes, history of MI, and smoking habits. A score is assigned to each level of the respective risk factor and the total score is calculated by adding up the points for each risk factor. The sum of these points provides the estimate of overall cardiovascular risk. For each total score, the corresponding absolute risk of MI or CHD death within the next 10 years is obtained [7]. In the PROCAM algorithm, the “high-risk” group had a 10-year coronary event risk greater than 20%, which corresponded to a PROCAM score of 53 and above. This definition of “high risk” of coronary event was reached through consensus by the Second Joint Task Force of European and other Societies on Coronary Prevention [9]. The American NCEP-ATP III guidelines [10] state that patients with an absolute risk of CHD exceeding 20% within the next 10 years should be considered at “high risk.” These patients require attention from health professionals and aggressive management of their modifiable risk factors.

The PROCAM score also estimates overall CHD risk in men aged 35 to 65 years. Regarding data on women, the limited number of coronary events occurring during the 10-year follow-up in the PROCAM study did not allow the development of a risk prediction algorithm specific to women. However, at that time, the PROCAM study did assess probability in women using the values of male subjects [7]. These analyses indicated that women aged 45 to 65 years have a fourfold lower absolute risk of coronary events compared to men of the same age. It has therefore been proposed that women’s risk could be estimated by dividing the global risk predicted for men by 4. However, several years later, investigators of the PROCAM study developed an updated CHD risk score that allowed the estimation of CHD risk in women as well as in elderly men and women [11]. The investigators also generated an algorithm to assess the 10-year risk of developing cerebral ischemic stroke in men and women aged 35-65 years [11]. An important feature of the PROCAM score is that it features only “hard” endpoints, including acute MI and sudden coronary death. The authors defined major coronary events as the primary coronary endpoints because these hard endpoints were unlikely to go unnoticed or be misdiagnosed [7].

Key Findings of the PROCAM Study

One of the most important findings of the PROCAM study was the recognition that risk factors do not act in isolation but in conjunction with other risk factors. Similar to previous prospective studies [12], individual risk factors were found to have a multiplicative rather than additive effect in predicting CHD events. An example of this is the interaction between total cholesterol and HDL cholesterol reported in the PROCAM study. After stratifying subjects into quartiles of cholesterol levels and tertiles of HDL cholesterol levels, CHD risk increased considerably (~60-fold) among men with elevated total cholesterol (>7.77 mmol/l or >300 mg/dl) and low HDL cholesterol (<0.91 mmol/l or <35 mg/dl) when compared to men with a total cholesterol below 5.18 mmol/l (200 mg/dl) and HDL cholesterol above 1.42 mmol/l (55 mg/dl) [13]. This interaction between total cholesterol and HDL cholesterol was the strongest predictor of CHD among men of the PROCAM study.

The PROCAM study also highlighted the importance of HDL cholesterol in the development of CHD [14]. Individuals with low HDL cholesterol levels had a roughly fourfold greater risk of CHD compared to individuals with normal HDL cholesterol levels. When stratified into tertiles of HDL cholesterol levels, the incidence of CHD among individuals in the first tertile (lowest HDL cholesterol levels) was higher than that of individuals in the top tertile (highest HDL cholesterol levels). Moreover, compared to men who remained asymptomatic, subjects with MI had lower HDL cholesterol distribution values.

In addition, the stratification of subjects for both HDL and LDL cholesterol levels revealed that individuals with low HDL cholesterol levels had an increased risk of CHD regardless of their LDL cholesterol concentration [14]. However, the worst-case scenario was the combination of low HDL cholesterol and elevated LDL cholesterol. This combination was clearly tied to the highest CHD risk, far beyond that of elevated LDL cholesterol or low HDL cholesterol on an individual basis. HDL cholesterol levels therefore have a much stronger influence on CHD risk in individuals with elevated global cardiovascular risk [15], supporting the notion that risk factors do not act in isolation but in synergy.

Multivariate analyses from the PROCAM study [13] also demonstrated that cholesterol, HDL cholesterol, LDL cholesterol, and triglycerides showed a significant age-adjusted association with incidence of acute coronary events. Moreover, the association between triglycerides and CHD risk remained significant even after adjusting for LDL cholesterol, HDL cholesterol, and other traditional risk factors such as age, systolic blood pressure, cigarette smoking, type 2 diabetes, family history of MI, and angina pectoris.

In addition, an analysis of the incidence of coronary events revealed that the highest PROCAM score quintile had more than a thirtyfold increased risk of coronary events compared to the lowest quintile [16]. These estimates were based on a Cox proportional hazard model derived from multiple metabolic risk variables (age, LDL cholesterol, HDL cholesterol, systolic blood pressure, triglycerides, cigarette smoking, type 2 diabetes, and family history of MI).

Several studies have explored to what extent the PROCAM score could apply to other populations. Cooper et al. [17] compared the predictive value of the PROCAM and Framingham risk algorithms in healthy men from the United Kingdom recruited for the Second Northwick Park Heart Study (NPHS-II) and followed for a median of 10.8 years for CHD events. Interestingly, despite the fact that the PROCAM score was derived from a European population (Germany), this algorithm was marginally better than the Framingham score in predicting CHD in men from the United Kingdom. Moreover, both the PROCAM and Framingham scores tended to overestimate the rates of MI and sudden coronary death. Similarly, Empana et al. [18] assessed the applicability of PROCAM and Framingham risk functions to middle-aged men from Northern Ireland and France in the Prospective Epidemiological Study of Myocardial Infraction (étude Prospective de l’Infarctus du MyocardE-PRIME) cohort. This study showed that both risk functions clearly overestimated the absolute CHD risk of men from Belfast and France. Therefore, while PROCAM risk functions may be suitable for classifying individuals according to their estimated absolute CHD risk, they may be inappropriate to estimate absolute CHD risk among healthy middle-aged men from low-risk (France) and high-risk (Belfast) populations, as this leads to an overestimation. These results highlight the need to perform additional cohort studies in various ethnic groups and parts of the world to provide risk assessment tools that are relevant to other populations.

In summary, the prospective PROCAM study has allowed to better understand key factors involved in the development of CHD and stroke. The risk algorithms generated over the years provide a simple, practical way to estimate CHD risk and stroke using commonly measured risk factors.

References

  1. Giampaoli S, Palmieri L, Mattiello A, et al. Definition of high risk individuals to optimise strategies for primary prevention of cardiovascular diseases. Nutr Metab Cardiovasc Dis 2005; 15: 79-85.

    PubMed ID: 15871855
  2. Greenland P, Knoll MD, Stamler J, et al. Major risk factors as antecedents of fatal and nonfatal coronary heart disease events. JAMA 2003; 290: 891-7.

    PubMed ID: 12928465
  3. Khot UN, Khot MB, Bajzer CT, et al. Prevalence of conventional risk factors in patients with coronary heart disease. JAMA 2003; 290: 898-904.

    PubMed ID: 12928466
  4. Kannel WB, Dawber TR, Friedman GD, et al. Risk factors in coronary heart disease. An evaluation of several serum lipids as predictors of coronary heart disease; The Framingham Study. Ann Intern Med 1964; 61: 888-99.

    PubMed ID: 14233810
  5. Wilson PW. Established risk factors and coronary artery disease: the Framingham Study. Am J Hypertens 1994; 7: 7S-12S.

    PubMed ID: 7946184
  6. Wilson PW, D’Agostino RB, Levy D, et al. Prediction of coronary heart disease using risk factor categories. Circulation 1998; 97: 1837-47.

    PubMed ID: 9603539
  7. Assmann G, Cullen P and Schulte H. Simple scoring scheme for calculating the risk of acute coronary events based on the 10-year follow-up of the prospective cardiovascular Munster (PROCAM) study. Circulation 2002; 105: 310-5.

    PubMed ID: 11804985
  8. Cullen P, Schulte H and Assmann G. Smoking, lipoproteins and coronary heart disease risk. Data from the Munster Heart Study (PROCAM). Eur Heart J 1998; 19: 1632-41.

    PubMed ID: 9857915
  9. Prevention of coronary heart disease in clinical practice. Recommendations of the Second Joint Task Force of European and other Societies on coronary prevention. Eur Heart J 1998; 19: 1434-503.

    PubMed ID: 9820987
  10. Expert Panel on Detection E, and Treatment of High Blood Cholesterol in Adults. Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). JAMA 2001; 285: 2486-97.

    PubMed ID: 11368702
  11. Assmann G, Schulte H, Cullen P, et al. Assessing risk of myocardial infarction and stroke: new data from the Prospective Cardiovascular Munster (PROCAM) study. Eur J Clin Invest 2007; 37: 925-32.

    PubMed ID: 18036028
  12. Grundy SM, Pasternak R, Greenland P, et al. Assessment of cardiovascular risk by use of multiple-risk-factor assessment equations: a statement for healthcare professionals from the American Heart Association and the American College of Cardiology. Circulation 1999; 100: 1481-92.

    PubMed ID: 10500053
  13. Assmann G, Cullen P and Schulte H. The Munster Heart Study (PROCAM). Results of follow-up at 8 years. Eur Heart J 1998; 19 Suppl A: A2-11.

    PubMed ID: 9519336
  14. Assmann G, Schulte H, von Eckardstein A, et al. High-density lipoprotein cholesterol as a predictor of coronary heart disease risk. The PROCAM experience and pathophysiological implications for reverse cholesterol transport. Atherosclerosis 1996; 124 Suppl: S11-20.

    PubMed ID: 8831911
  15. Assmann G. Calculating global risk: the key to intervention. Eur Heart J Suppl 2005; 7: F9-F14. https://academic.oup.com/eurheartjsupp/article/7/suppl_F/F9/578417. Accessed October 19, 2020.

    PubMed ID:
  16. Assmann G, Carmena R, Cullen P, et al. Coronary heart disease: Reducing the risk. A worldwide view. Circulation 1999; 100: 1930-1938.

    PubMed ID: 10545439
  17. Cooper JA, Miller GJ and Humphries SE. A comparison of the PROCAM and Framingham point-scoring systems for estimation of individual risk of coronary heart disease in the Second Northwick Park Heart Study. Atherosclerosis 2005; 181: 93-100.

    PubMed ID: 15939059
  18. Empana JP, Ducimetiere P, Arveiler D, et al. Are the Framingham and PROCAM coronary heart disease risk functions applicable to different European populations? The PRIME Study. Eur Heart J 2003; 24: 1903-11.

    PubMed ID: 14585248
Reference 1 CLOSECLOSE

Giampaoli S, Palmieri L, Mattiello A, et al. Definition of high risk individuals to optimise strategies for primary prevention of cardiovascular diseases. Nutr Metab Cardiovasc Dis 2005; 15: 79-85.

PubMed ID: 15871855
Reference 2 CLOSECLOSE

Greenland P, Knoll MD, Stamler J, et al. Major risk factors as antecedents of fatal and nonfatal coronary heart disease events. JAMA 2003; 290: 891-7.

PubMed ID: 12928465
Reference 3 CLOSECLOSE

Khot UN, Khot MB, Bajzer CT, et al. Prevalence of conventional risk factors in patients with coronary heart disease. JAMA 2003; 290: 898-904.

PubMed ID: 12928466
Reference 4 CLOSECLOSE

Kannel WB, Dawber TR, Friedman GD, et al. Risk factors in coronary heart disease. An evaluation of several serum lipids as predictors of coronary heart disease; The Framingham Study. Ann Intern Med 1964; 61: 888-99.

PubMed ID: 14233810
Reference 5 CLOSECLOSE

Wilson PW. Established risk factors and coronary artery disease: the Framingham Study. Am J Hypertens 1994; 7: 7S-12S.

PubMed ID: 7946184
Reference 6 CLOSECLOSE

Wilson PW, D’Agostino RB, Levy D, et al. Prediction of coronary heart disease using risk factor categories. Circulation 1998; 97: 1837-47.

PubMed ID: 9603539
Reference 7 CLOSECLOSE

Assmann G, Cullen P and Schulte H. Simple scoring scheme for calculating the risk of acute coronary events based on the 10-year follow-up of the prospective cardiovascular Munster (PROCAM) study. Circulation 2002; 105: 310-5.

PubMed ID: 11804985
Reference 8 CLOSECLOSE

Cullen P, Schulte H and Assmann G. Smoking, lipoproteins and coronary heart disease risk. Data from the Munster Heart Study (PROCAM). Eur Heart J 1998; 19: 1632-41.

PubMed ID: 9857915
Reference 9 CLOSECLOSE

Prevention of coronary heart disease in clinical practice. Recommendations of the Second Joint Task Force of European and other Societies on coronary prevention. Eur Heart J 1998; 19: 1434-503.

PubMed ID: 9820987
Reference 10 CLOSECLOSE

Expert Panel on Detection E, and Treatment of High Blood Cholesterol in Adults. Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). JAMA 2001; 285: 2486-97.

PubMed ID: 11368702
Reference 11 CLOSECLOSE

Assmann G, Schulte H, Cullen P, et al. Assessing risk of myocardial infarction and stroke: new data from the Prospective Cardiovascular Munster (PROCAM) study. Eur J Clin Invest 2007; 37: 925-32.

PubMed ID: 18036028
Reference 12 CLOSECLOSE

Grundy SM, Pasternak R, Greenland P, et al. Assessment of cardiovascular risk by use of multiple-risk-factor assessment equations: a statement for healthcare professionals from the American Heart Association and the American College of Cardiology. Circulation 1999; 100: 1481-92.

PubMed ID: 10500053
Reference 13 CLOSECLOSE

Assmann G, Cullen P and Schulte H. The Munster Heart Study (PROCAM). Results of follow-up at 8 years. Eur Heart J 1998; 19 Suppl A: A2-11.

PubMed ID: 9519336
Reference 14 CLOSECLOSE

Assmann G, Schulte H, von Eckardstein A, et al. High-density lipoprotein cholesterol as a predictor of coronary heart disease risk. The PROCAM experience and pathophysiological implications for reverse cholesterol transport. Atherosclerosis 1996; 124 Suppl: S11-20.

PubMed ID: 8831911
Reference 15 CLOSECLOSE

Assmann G. Calculating global risk: the key to intervention. Eur Heart J Suppl 2005; 7: F9-F14. https://academic.oup.com/eurheartjsupp/article/7/suppl_F/F9/578417. Accessed October 19, 2020.

PubMed ID:
Reference 16 CLOSECLOSE

Assmann G, Carmena R, Cullen P, et al. Coronary heart disease: Reducing the risk. A worldwide view. Circulation 1999; 100: 1930-1938.

PubMed ID: 10545439
Reference 17 CLOSECLOSE

Cooper JA, Miller GJ and Humphries SE. A comparison of the PROCAM and Framingham point-scoring systems for estimation of individual risk of coronary heart disease in the Second Northwick Park Heart Study. Atherosclerosis 2005; 181: 93-100.

PubMed ID: 15939059
Reference 18 CLOSECLOSE

Empana JP, Ducimetiere P, Arveiler D, et al. Are the Framingham and PROCAM coronary heart disease risk functions applicable to different European populations? The PRIME Study. Eur Heart J 2003; 24: 1903-11.

PubMed ID: 14585248