Relationship Between Cardiovascular Risk as Predicted by Established Risk Scores Versus Plaque Progression as Measured by Serial Intravascular Ultrasound in Left Main Coronary Arteries
Background— Intravascular ultrasound (IVUS) is increasingly used as an end point in studies aimed at reducing progression or inducing regression of coronary artery disease. However, data linking serial changes by IVUS with clinical outcomes are scarce.
Methods and Results— In the absence of a validated risk score for secondary prevention, we compared 3 established risk scores for primary prevention—PROCAM, SCORE, and Framingham—with plaque progression and lumen reduction as assessed with serial IVUS (follow-up, 18±9 months) in atherosclerotic left main coronary arteries of 56 patients with established atherosclerosis. For all 3 algorithms, patients at highest estimated risk of events showed greater plaque progression than patients at lowest risk (P<0.05 to <0.01). There were positive linear relationships between the risk of clinical events and plaque progression (r=0.41 to 0.60; P<0.002 to <0.0001). This translated into a greater decrease in lumen dimensions with increasing risk (P<0.05, PROCAM and SCORE). Risk prediction using the PROCAM algorithm showed the strongest relation with serial IVUS. During follow-up, 18 patients suffered from adverse cardiovascular events; these patients had an annual plaque progression that was significantly greater than other patients (25.2±19.4% versus 5.9±15.6%, P<0.001).
Conclusions— There was a positive linear relationship between the estimated risk of clinical events derived from all 3 established risk-score algorithms and the extent of plaque progression measured by serial IVUS. This translated into stenosis progression (reduction in lumen dimensions) with increasing clinical risk.
Received January 16, 2004; de novo received February 26, 2004; revision received May 12, 2004; accepted May 19, 2004.
Coronary artery disease is a major cause of death and disability in Western-lifestyle countries. Intravascular ultrasound (IVUS) permits measurements of both lumen and plaque dimensions in vivo.1 Serial IVUS studies can be used to assess the progression of atherosclerotic plaque dimensions in native coronary arteries.2–6 Several pharmacological intervention studies with clinical end points have demonstrated improved patient outcomes with risk factor modification.7–10 In addition, progression-regression studies with angiographic end points have demonstrated the relationship between angiographic disease progression and clinical outcome.11 However, thus far, there is a lack of data linking changes in plaque dimensions as measured with IVUS with clinical outcome.5 Guidelines on prevention of coronary heart disease use predicted 10-year risk of coronary events to identify candidates for risk factor modification.12–15 Ideally, a predictor of the overall risk of clinical events should be based on a multifactorial model.16,17 Published “primary prevention” risk-scoring methods use equations derived from large cohort studies, such as the recent European Systematic COronary Risk Evaluation (SCORE) Project,16 the German PROspective CArdiovascular Münster (PROCAM) Study,18 and the US Framingham Heart and Offspring Studies.19 In the absence of a validated risk score for secondary prevention, we used these established primary-event risk scores17–19 to assess the relationship between the estimated risk of clinical events and plaque progression as assessed with serial IVUS in atherosclerotic left main stems of patients with symptomatic coronary artery disease.
We compared established risk scores with serial IVUS studies of left coronary atherosclerotic plaques in a population that has previously been reported.6 All plaques were de novo, were hemodynamically nonsignificant, and met the following criteria: (1) serial high-quality IVUS of the entire left main artery ≥12 months apart; (2) calcifications that did not limit quantitative assessment of vessel cross-sectional area (CSA); (3) nonostial plaque location; (4) angiographic lumen diameter stenosis <30% (“worst-view” visual assessment); and (5) no intervention in the very proximal left anterior descending or circumflex coronary arteries because these interventions could have affected the left main artery. Of these cases (n=60), the present analysis includes the data of 56 patients <65 years old, because most risk scores are validated only for patients up to that age. Patients were examined in the Essen University Cardiac Catheterization Laboratory, with a follow-up of 18±9 months (median, 14 months; range, 12 to 50 months). The Local Council on Human Research approved this IVUS study, and all patients signed a written informed consent form as approved by the Local Medical Ethics Committee. As previously reported, this represents a consecutive series of patients who underwent initial IVUS examination during coronary intervention and then returned after ≥1 year for repeat intervention, during which another IVUS study was performed.6
Cardiovascular Risk Factors, Parameters, and Medication
In our laboratory, we prospectively record demographics, cardiovascular risk factors, medications, and results of key laboratory tests of patients examined with IVUS. Baseline laboratory tests were performed in all patients treated by percutaneous interventions and were analyzed in the central laboratory of Essen University according to international standards. Medications were recorded only if drugs were taken for >50% of the follow-up interval (eg, clopidogrel for 4 weeks was not tabulated). Tabulated cardiovascular risk factors and parameters included sex, age, systolic blood pressure, serum total cholesterol level, serum LDL cholesterol level, serum HDL cholesterol level, serum triglyceride level, history of smoking during the previous 12 months (and also the previous month), family history of coronary artery disease (myocardial infarction of first-degree relative <60 years of age), and diabetes mellitus (known diabetes or repeated fasting blood glucose levels >120 mg/dL). Plasma concentrations of total cholesterol, LDL cholesterol, HDL cholesterol, and triglycerides were measured by standard enzymatic methods.
Cardiovascular Risk Assessment by PROCAM Score
The PROCAM score was calculated as previously described.18 It was derived from the European PROCAM study, performed in a population of ≈5000 participants in Münster, Germany. The score considers sex, age, LDL cholesterol, HDL cholesterol, triglycerides, systolic blood pressure, smoking, diabetes mellitus, and family history. The categories for continuous variables of age, systolic blood pressure, LDL cholesterol, and HDL cholesterol were based on the National Cholesterol Education Program III guidelines15; the categories for triglycerides were based on the guidelines of the International Task Force for Prevention of Coronary Heart Disease.20 For each patient, 10-year risk of coronary events (fatal/nonfatal myocardial infarction or sudden death) was predicted.18
Cardiovascular Risk by SCORE Risk Assessment System
The SCORE risk was determined from a risk assessment algorithm previously described.16 The SCORE risk assessment system considers sex, age, total cholesterol, systolic blood pressure, and smoking. We used the table for populations at high cardiovascular disease risk on the basis of total/HDL cholesterol levels. It is derived from a large (>205 000 participants) data set of prospective European studies. It predicted the 10-year risk of fatal cardiovascular events.16
Cardiovascular Risk Assessment by Framingham Score
The Framingham score was calculated by use of an algorithm previously described.15 The score considers sex, age, total cholesterol, HDL cholesterol, systolic blood pressure, and smoking. The Framingham score is based on data from a sample of the Framingham Heart and Offspring studies.19 It was used to predict the 10-year risk of coronary events (fatal/nonfatal myocardial infarction or sudden death).
IVUS imaging was performed as previously described.6 In brief, IVUS studies were performed during percutaneous coronary interventions of mid or distal left anterior descending or circumflex arteries after intracoronary injections of 200 μg nitroglycerin with commercially available IVUS systems: a mechanical sector scanner (Boston Scientific Corp) incorporating a 30-MHz single-element beveled transducer or a solid-state device (Endosonics). Importantly, at Essen University, if a patient undergoes imaging with one IVUS system during an index procedure, the same IVUS system is used at follow-up. Slow, continuous pullbacks were generally performed with a motorized pullback device (at 0.5 mm/s). In addition, a dedicated image-in-image system (Echo-Map, Siemens)21 was used to record the “angiographic” position of the IVUS probe together with the corresponding IVUS image, especially at sites of characteristic landmarks (ie, calcifications or unusual plaque shapes) and/or the target site. Follow-up IVUS studies were performed (using the same IVUS system as at the initial measurement) during repeat coronary interventions and during IVUS examinations of ambiguous coronary lesions or (clinically driven) follow-up catheterizations.
Quantitative IVUS Analysis
The left main artery target site image slice was determined from the initial IVUS study; this was the site with the smallest lumen CSA within the plaque.6 If there were several slices with equal lumen size, the one with the largest external elastic membrane (EEM) and plaque plus media (P&M=EEM minus lumen) CSA was analyzed.1,22,23 Exact matching of the target site on initial and follow-up IVUS studies was ensured by use of side-by-side comparison of the 2 IVUS sequences along with the pullback speed; the operators’ recorded comments (on video tape); and characteristic calcifications, vascular and perivascular landmarks, and plaque shapes. If required, x-ray sequences of the dedicated image-in-image system (Echo-Map) were revisited to optimize matching.21
The lumen CSA was measured by tracing the leading edge of the intima. The EEM CSA was measured by tracing the leading edge of the adventitia. Extrapolation of the EEM boundary behind calcium was possible if each individual calcific deposit did not shadow >75° of the adventitial circumference. In our laboratory, the intraclass correlation coefficient is 0.99 for repeated measurements of EEM, 0.96 for lumen, and 0.99 for P&M CSA. To compensate for variations in follow-up intervals and to obtain comparable data, we calculated absolute and relative changes (Δ) between initial and follow-up IVUS data; measurements were normalized for the length of the follow-up period (changes per year) and baseline variables.
To evaluate the potential effect of individual risk factors on annual plaque progression, we compared changes in P&M CSA of patients with high versus low values of individual risk factors. For noncontinuous variables, we compared patients with presence versus absence of individual risk factors; for continuous variables, we applied the (rounded) median of that parameter to create groups (value <median versus value ≥median). The low number of female patients (n=6) did not justify further testing for sex differences.
Analyses were performed with SPSS 10.0.7 (Microsoft) for Windows. Dichotomous data are presented as frequencies. Quantitative data are presented as mean±SD and compared using the Student t test, regression analysis, or ANOVA for repeated measures with post hoc testing with the Tukey honestly significant difference test. A probability value of P<0.05 was considered significant.
Demographics, Lipid Profile, and Medication
Demographics, lipid profile, and medication of the study population are presented in Table 1.
Estimated Risk of Clinical Events and Serial IVUS Data
For the 3 risk algorithms (PROCAM, SCORE, and Framingham), the changes in P&M, lumen, and EEM CSA for patients with <10%, 10% to 20%, and >20% risk of clinical events are given in Tables 2 through 4⇓⇓. For all 3 risk algorithms, patients at highest estimated risk showed the greatest IVUS plaque progression.
Individual Risk Factors, Medication, and Plaque Progression
Annual changes in P&M CSA of patients with high versus low values of individual risk factors are shown in Figure 1; P&M CSA progression was significantly greater for smokers and for patients with unfavorable serum cholesterol profile. When studied as continuous variables, annual changes in P&M CSA correlated with LDL and HDL cholesterol as previously reported6; with total cholesterol (r=0.27, P<0.05) and total cholesterol/HDL ratio (r=0.43, P<0.001) but not with triglycerides (r=−0.02, P=0.9); and not with age (r=0.06, P=0.7) or systolic blood pressure (r=0.07, P=0.6). Statin use (n=47) was associated with less P&M CSA progression (11.2±19.6% versus 24.2±18.6%, P<0.05); other drugs had no impact.
Estimated Risk Versus Changes in Plaque and Lumen Dimensions
The relation between changes in P&M CSA and lumen CSA versus estimated risk of events is shown in Figure 2. There was a linear relation between P&M CSA progression versus estimated risk of clinical events and between lumen CSA decrease versus estimated risk of clinical events.
Plaque Progression and Actual Adverse Cardiovascular Events
During follow-up, 5 patients had an acute myocardial infarction; 7, unstable angina; and 6, percutaneous interventions of new de novo coronary lesions. Figure 3 presents annual P&M CSA changes together with information on adverse events for each patient.
Patients with adverse events (n=18) had more annual P&M CSA progression than the rest of the population (25.2±19.4% versus 5.9±15.6%, P<0.001) but showed no significant difference in length of follow-up. Similarly, patients with myocardial infarction (n=5) as well as patients with myocardial infarction or unstable angina (n=12) had more annual P&M CSA progression than the rest of the population (30.9±7.4% versus 9.9±18.7%, P<0.001, and 29.8±15.3% versus 6.9±16.8%, P<0.0003, respectively).
This study showed a positive linear relationship between estimated risk of clinical events derived from all 3 established risk scores and the extent of plaque progression as measured with serial IVUS. This translated into a greater decrease in lumen dimensions with increasing risk of clinical events. For all 3 algorithms, patients with the highest estimated risk of clinical events showed the greatest plaque progression by IVUS. Risk prediction by PROCAM score showed the strongest correlation with serial IVUS measurements.
Our findings are of interest because there is an increasing use of IVUS as an end point of pharmacological interventions designed to limit progression or induce regression of coronary artery disease.2–6 However, despite the present results, there is a lack of data linking changes in plaque dimensions, as measured with IVUS, with clinical outcome.5 Thus, there is a need for future prospective trials to combine serial IVUS assessment of coronary plaque progression with long-term clinical follow-up.
Risk Assessment in Patients With Established Coronary Heart Disease
In this study, we compared plaque progression by serial IVUS in patients with established coronary artery disease, a secondary measure, with established risk scores for primary disease prevention. We consider this approach to be reasonable in the absence of a validated risk score for secondary prevention. Although the major risk factors that are important in primary prevention remain important in secondary prevention,12,24–26 all individuals examined in the present study have an absolute risk of clinical events that was higher than suggested by the nominal results of the 3 algorithms. Nevertheless, we were able to demonstrate with all 3 algorithms a linear relationship between estimated risk of clinical events and the extent of atherosclerotic plaque progression as measured with serial IVUS.
The risk-scoring methods studied in the present report were derived from the European SCORE Project,16 the German PROCAM Study,18 and the US Framingham Heart and Offspring Studies.19 In fact, the PROCAM Study was performed in the city of Münster; Münster is located next to Essen, the city in which the present IVUS study was performed. Therefore, perhaps not surprisingly, we observed a particularly strong relationship between plaque progression and estimated risk of clinical events using the PROCAM score. Conversely, the correlation of the SCORE algorithm and (even more) of the Framingham risk score showed relationships with the current IVUS data that were less strong but still significant. This suggests that risk scores derived from population studies may have a somewhat limited applicability when applied to another population with a different risk profile from a different geographical region, even if both populations were Western communities.27 This may be even more of a problem when the 2 populations are from very different cultures and genetic makeups, for example, applying Western scores to Asian patients.
Of note, SCORE, PROCAM, and Framingham algorithms used different end points (risk of cardiac events versus cardiovascular mortality; fatal versus combined fatal and nonfatal end points); this makes direct comparison difficult. Nevertheless, our findings were quite similar among the 3 risk scores.
Our data suggest that compared with conventional studies of cardiovascular risk, IVUS assessment of plaque and lumen dimensions may provide valuable insights from relatively small study populations within a relatively short period of time. This article may be a first step to applying direct assessment of plaque progression or regression by IVUS to fine-tune our understanding of cardiovascular risk factors. As a consequence, IVUS assessment of coronary atherosclerosis may be used to develop and/or validate secondary risk factor scores in the future.
By most standards, this was a large serial IVUS study; however, all studies with long-term serial IVUS assessment of atherosclerosis are limited to a relatively small number of patients. We were able to include only patients with significant coronary artery disease who were admitted for repeat cardiac catheterization ≥12 months after baseline. Because of this selection bias, the findings of the present study may not be applicable to a population of patients treated by a single percutaneous coronary intervention or even to the general population. We used 2 IVUS systems in the present study; although this approach may have minor shortcomings, every effort was made to obtain the most reliable data possible, as discussed in detail previously.6 Three-dimensional IVUS analysis may be superior for the assessment of progression/regression. The patients in the present study had established coronary atherosclerosis. We studied only left main disease as representative of segments that did not receive intervention.
There is a positive linear relation between the estimated risk of clinical events derived from 3 established risk scores (that were initially developed to predict primary events) and the extent of plaque progression as measured with serial IVUS in patients with established coronary artery disease. For all 3 algorithms, patients at highest estimated risk showed the greatest plaque progression by IVUS. This translated into the greatest reduction in lumen dimensions in patients with the highest risk of events.
Schartl M, Bocksch W, Koschyk DH, et al, for the GAIN-Study Investigators. Use of intravascular ultrasound to compare effects of different strategies of lipid-lowering therapy on plaque volume and composition in patients with coronary artery disease. Circulation. 2001; 104: 387–392.
von Birgelen C, Hartmann M, Mintz GS, et al. Relation between progression and regression of atherosclerotic left main coronary artery disease and serum cholesterol levels as assessed with serial long-term (≥12 months) follow-up intravascular ultrasound. Circulation. 2003; 108: 2757–2762.
Scandinavian Simvastatin Survival Study Group. Randomized trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet. 1994; 19:344: 1383–1389.
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: 1481–1492.
Fodor JG, Frohlich JJ, Genest JJ Jr, et al. Recommendations for the management and treatment of dyslipidemia: report of the working group on hypercholesterolemia and other dyslipidemias. CMAJ. 2000; 162: 1441–1447.
Expert Panel on Detection, Evaluation, 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: 2486–2497.
Conroy RM, Pyörälä K, Fitzgerald AP, et al, for the SCORE project group. Estimation of ten-year risk of fatal cardiovascular disease in Europe: the SCORE project. Eur Heart J. 2003; 24: 987–1003.
De Backer G, Ambrosioni E, Borch-Johnsen K, et al. European guidelines on cardiovascular disease prevention in clinical practice. Eur Heart J. 2003; 24: 1601–1610.
Assmann G, Cullen P, Schulte H. Simple scoring scheme for calculating the risk of acute coronary events based on the 10-year follow-up of the Prospective Cardiovascular Münster (PROCAM) Study [published correction appears in Circulation. 2002;105:900]. Circulation. 2002; 105: 310–315.
Anderson KM, Wilson PW, Odell PM, et al. An updated coronary risk profile: a statement for health professionals. Circulation. 1991; 83: 356–362.
Assmann G, Carmena R, Cullen P, et al. Coronary heart disease: reducing the risk: a worldwide view. International Task Force for the Prevention of Coronary Heart Disease. Circulation. 1999; 100: 1930–1938.
Genest J, Pedersen TR. Prevention of cardiovascular ischemic events: high-risk and secondary prevention. Circulation. 2003; 107: 2059–2065.
Hense HW, Schulte H, Lowel H, et al. Framingham risk function overestimates risk of coronary heart disease in men and women from Germany: results from MONICA Augsburg and the PROCAM cohorts. Eur Heart J. 2003; 24: 1–9.