The Relation between Extent of Coronary Artery Disease<br />
Measured by Quantitative Coronary Angiography and Changes<br />
in Lipid Prof

Coronary artery disease (CAD) is inexorably linked to abnormal cholesterol levels, particularly low-density lipoprotein- C (LDL-C).^1–4 Longstanding hyperlipidemia leads to deposition of oxidized LDL-C in activated macrophages and results, ultimately, in the formation of the atherosclerotic plaque. As it became apparent in the last three decades, most of the plaque resides in the coronary artery wall. The remodeling associated with this process enables patients to have large volumes of plaque without appreciable reduction in the dimensions of the coronary lumen available for blood flow.⁵
Since the introduction of HMG Co-A reductase inhibitors (statins), dramatic reductions in the incidence of CAD events were observed.⁶ Yet, commensurate changes in the degree of coronary obstructions have been more difficult to observe, even with the use of quantitative coronary angiography (QCA). The “uncoupling” of the dramatic effect on clinical events from changes in coronary lumen dimensions has raised doubts about the utility of QCA as a surrogate endpoint of anti-atherosclerotic therapy success. Alternative methods of coronary imaging, such as intravascular ultrasound (IVUS), have become popular and were deemed more likely to represent the true anatomical changes in plaque size and distribution.^7–9
We set out to characterize changes in QCA-determined measures of CAD in response to changes in LDL-C and highdensity lipoprotein (HDL) in patients with serial angiographic studies enrolled in studies of atherosclerosis regression.

Methods
Our angiographic core laboratory performed QCA in patients enrolled in three trials of atherosclerosis regression based on lipid-modifying therapy, which have been reported in detail. In brief, the 1,315 patients in this study were enrolled in: A Study to Evaluate the Effect of Rosuvastatin on Intravascular Ultrasound-Derived Coronary Atheroma Burden (ASTEROID);¹⁰ the Reversal of Atherosclerosis with Aggressive Lipid Lowering [REVERSAL] trial;¹¹ and the ACAT Intravascular Atherosclerosis Treatment Evaluation (ACTIVATE) trial.¹² In ASTEROID, patients were treated in an open-label registry with rosuvastatin, a potent statin affecting HDL to a greater extent than other statins. In REVERSAL, patients were randomized to pravastatin 40 mg or atorvastatin 80 mg daily in a double-blind, randomized trial. In ACTIVATE, patients were randomized to the enzyme acyl–coenzyme A:cholesterol acyltransferase (ACAT) inhibitor or matching placebo in addition to regular preventive therapy. Lipid profiles were available for all patients upon entry in the trial and at protocol-mandated intervals during the study (3–6 months). The follow-up level is the average of the measurements during the study after the initiation of therapy. Similarly, CRP levels were available in 909 patients. Only the patients with adequate angiographic studies, performed 18–24 months apart, were included in this report. For each patient, 10 segments (3 in the right coronary artery, 1 in the left main coronary artery, 3 in the left anterior descending artery and 3 in the left circumflex artery) were analyzed at baseline and follow up. Segments in vessels undergoing angioplasty at baseline or during the study were excluded from analysis at the end-of-study angiogram.
The methods for QCA were previously reported.¹³ Pie Medical Imaging (Maastricht, The Netherlands) QCA software was utilized. Automated contour detection was verified and corrected visually, if inadequate. Each angiogram was reviewed independently by experienced technicians and the medical director (SJB). In a quality assurance project for one of the angiographic analyses performed in our laboratory,¹⁴ the intraobserver correlation after blinded review for reference diameter (RD) and minimal lumen diameter (MLD) was 0.996 and 0.997, respectively (68 segments). In another trial of atherosclerosis reduction, the intraobserver correlation for RD was 0.976 and 0.970 for MLD in 114 segments.¹¹ For this study we analyzed three parameters of CAD extent: 1) CAD score, defined as the average MLD for all segments evaluated;^15,16 2) cumulative stenosis score, defined as the sum of the % stenosis (expressed as a number between 0 and 1) in the evaluated segments;¹⁷ and 3) average plaque area, defined as the average of the differences between the lumen area at the RD point and the lumen area at the MLD point.¹⁸ The change in each parameter was indexed per year of follow up to account for different intervals between angiograms and was analyzed against change (follow-up value – baseline value) in LDL, % change in LDL, and % change in HDL – % change in LDL. The latter was recently reported to correlate with angiographic progression of CAD.19 For comparison, we also provided the intravascular ultrasound (IVUS)-derived percent atheroma volume (PAV) for a selected segment in each patient, as previously described in detail.¹¹
Statistical analysis. Continuous and categorical values were compared with the appropriate nonparametric test or the chi-square test. A linear mixed regression model was applied for the univariate association between changes in QCA and lipid profile parameters. The level of significance was set at p < 0.05.

Results

The baseline characteristics of the study population are shown in Table 1. These were typical CAD patients in their sixth decade of life and with a high incidence of risk factors for CAD. A total of 6,165 segments were analyzed at baseline and follow up, or almost 5 segments per patient. The changes in angiographic parameters and lipid profile and CRP during the studies’ duration are shown in Table 2. Notably, the baseline average LDL was only modestly elevated at 128 ± 39 mg/dl, while the HDL was 43 ± 11 mg/dl. The coronary stenosis score indicates that the majority of segments had stenoses in the 20% range both at baseline and at follow up. There were no statistically significant changes in CAD score or stenosis score from baseline to follow-up study (p = 0.94 and 0.21, respectively). In contrast, angiographic average plaque area (available only in ACTIVATE and ASTEROID, n = 784) increased significantly (p < 0.001) and PAV increased insignificantly in the segments chosen for IVUS evaluation (p = 0.16).
The relation between absolute (Figure 1) or relative changes in LDL and CAD score are described by the linear equations: change of average MLD (coronary artery score) per year = -0.00337 – 0.00008 * change in LDL; p = 0.30 and change of average MLD (coronary artery score) per year = -0.00178 – 0.00007 * % change in LDL; p = 0.53. Similar results were obtained for the cumulative stenosis score (p = 0.20 and p = 0.10, respectively). There was no statistical correlation between the change in average plaque area per year and tbe change in LDL (p = 0.76) or % change in LDL (p = 0.45).

Changes in CAD score (Figure 2) and cumulative stenosis score also did not correlate with the summation of changes in HDL and LDL from baseline (% change in HDL – % change in LDL), as an index of beneficial change in lipid profile (p = 0.80 and p = 0.35, respectively). The relation between change in mean % stenosis (another parameter of CAD extent)19 and summation of changes in HDL and LDL is described by the equation: change in mean of % stenosis per year = 0.1426 – 0.00353 * (% change in HDL – % change in LDL); p = 0.10.
We subsequently divided the patients into groups of those who demonstrated regression in CAD (higher CAD score at follow up, n = 756) and those with progression or no change in CAD (n = 555). While baseline lipid profile was more favorable in the “regressors” (lower LDL and higher HDL, p = 0.01 for both), both groups had similar follow-up values (p = 0.12 and p = 0.38, respectively). CRP levels were similar in both groups both at baseline and at follow up. PAV was nearly 40% both at baseline and at follow up in both groups, and there was no difference in the change in these parameters between angiographic “regressors” and “progressors” (p = 0.93, Table 3).

Discussion
The main findings observed in this large cohort of patients treated with different lipid-modifying agents in three contemporary trials can be summarized as follows: 1) Annualized changes in the CAD score, representing the average MLD of segments with baseline and follow-up evaluation, were not significantly associated with absolute or relative changes in LDL values; 2) Annualized changes in the stenosis score, representing the sum of % stenoses of all segments with baseline and follow-up evaluation, were not significantly associated with absolute or relative changes in LDL values; 3) Even when changes in LDL and HDL were considered together, there was no significant association with changes in coronary score or stenosis score; 4) While there was no significant change in lumen dimensions, the increase in average plaque area (measured by angiography) and percent atheroma volume (measured by IVUS) suggest a certain degree of positive remodeling. This finding reinforces the concept that plaque growth cannot be optimally assessed with angiography and that vascular remodeling can accommodate plaque growth without significantly affecting area available for flow; and 5) There were no substantial differences between patients with progression in CAD score and those with regression in terms of levels of LDL, HDL or CRP achieved on therapy, even though, paradoxically, the “progressors” had a greater reduction in LDL and a greater increase in HDL than “regressors”, as they started with a less favorable lipid profile.
It is important to note that all segments belonging to vessels that underwent angioplasty between baseline and follow-up studies were excluded from analysis. Thus, our findings reflect mostly the angiographic behavior of coronary segments with moderate narrowing at baseline in patients with stable coronary disease, and should not be extrapolated to segments with severe disease at initial presentation or to patients with acute coronary syndromes.
Our findings are at odds with those reported by Brown et al in a meta-analysis of 23 trials of lipid-lowering therapy, encompassing nearly 83,000 patients.¹⁹ Nearly 4,000 of them underwent QCA at baseline and at 2–4 years after randomization. Using a different QCA parameter than what was used in this study, namely the mean % stenosis of the 9 proximal segments of the coronary tree, they reported a significant linear association with the total % change in LDL and HDL. For patients with a change > 50% in HDL and LDL, there was actually regression in the mean % stenosis. As can be appreciated visually in Figure 2, a significant number of our patients achieved this level of lipid modification without a significant change in the sum of segment stenoses. Even when we used their QCA parameter, we could not detect a significant relationship to changes in HDL and LDL. Berry at al recently reported their findings from the Avasimibe and progression of Coronary Lesions Assessed by Intravascular Ultrasound (APLUS) trial.²⁰ In 525 patients treated with an ACAT inhibitor, 88% were on statin therapy as well. Using similar QCA parameters to ours, there was no significant correlation between changes in coronary score or stenosis score and changes in IVUS-derived plaque volume. While their study did not address the relationship between changes in QCA and changes in lipids, it reinforced the dissociation between what we measure with QCA and what we measure with IVUS.^21,22
Despite these conflicting data, what is indisputable is the strong link between reduction in clinical events and changes in LDL^23,24 or HDL.²⁵ If, as claimed by Brown et al,¹⁹ changes in QCA parameters are related to changes in LDL and HDL, one can make the assumption that changes in QCA measurements will be representative of, and a surrogate for, a reduction in clinical events. Alternatively, if our findings better represent the complex relationship between changes in QCA parameters and alterations in lipid profile, regulatory agencies and clinical investigators need to reassess the use of QCA, or at least the lumen dimensions derived from it, as a surrogate endpoint for clinical events in trials testing drugs that modify the lipid profile. We believe that our findings capture the complexity of the issues related to imaging of CAD burden and provide a cautionary note for the enthusiastic embrace of results suggesting that very small changes in imaging parameters can really be assumed to mean “regression” of CAD. Frequently, changes in % stenosis of 1 or 2% are well within the range of measurement error, as are similar changes in plaque volume over a long segment of the coronary artery.
Study limitations. We recognize several limitations in our findings. While we report on the QCA and lipid profile of the patients enrolled in three contemporary trials, we do not present data on clinical outcomes because the number of events in these studies was extremely low, reflecting a low-risk patient population with stable CAD followed for a relatively short time period. Even though the time period was short in this study, in trials of secondary prevention of CAD, the survival curves started to separate in the 12- to 24-month range.^24,26 One would expect to already observe changes in lumen dimensions if, indeed, the reduction in events were linked to substantial changes in lumen size. The exclusion of segments from vessels that underwent angioplasty essentially selects relatively stable segments of the coronary tree with little obstruction at baseline. Since the CAD score and stenosis score integrate the worst point of each segment at two timepoints, it is impossible to verify that these points are the same at the two examinations.

Conclusion
Despite these methodological limitations and acknowledging the inherent limitations of QCA with respect to evaluation of the vessel lumen only, we conclude that global parameters of coronary disease burden measured by QCA in low-risk patients followed for 2 years do not correlate significantly with changes in LDL, HDL or their combined changes. These findings, particularly if confirmed prospectively in adequately-sized trials, should prompt reevaluation of the use of surrogate imaging endpoints for predicting the effect of lipid-altering strategies on clinical outcomes.

References

1. Libby P. Changing concepts of atherogenesis. J Intern Med 2000;247:349–358.
2. Libby P, Ridker PM, Maseri A. Inflammation and atherosclerosis. Circulation 2002;105:1135–1043.
3. Libby P, Theroux P. Pathophysiology of coronary artery disease. Circulation 2005;111:3481–3488.
4. Ross R. Atherosclerosis — An inflammatory disease. N Engl J Med 1999;340:115–126.
5. Glagov S, Weisenberg E, Zarins CK, et al. Compensatory enlargement of human atherosclerotic coronary arteries. N Engl J Med 1987;316:1371–1375.
6. Ross SD, Allen IE, Connelly JE, et al. Clinical outcomes in statin treatment trials: A meta-analysis. Arch Intern Med 1999;159:1793–1802.
7. Nissen SE, Yock P. Intravascular ultrasound: Novel pathophysiological insights and current clinical applications. Circulation 2001;103:604–616.
8. Schartl M, Bocksch W, Koschyk DH, et al. 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.
9. Nissen S. Coronary angiography and intravascular ultrasound. Am J Cardiol 2001;87:15A–20A.
10. Nissen SE, Nicholls SJ, Sipahi I, et al. Effect of very high-intensity statin therapy on regression of coronary atherosclerosis: The ASTEROID trial. JAMA 2006;295:1556–1565.
11. Nissen SE, Tuzcu EM, Schoenhagen P, et al. Effect of intensive compared with moderate lipid-lowering therapy on progression of coronary atherosclerosis: A randomized controlled trial. JAMA 2004;291:1071–1080.
12. Nissen SE, Tuzcu EM, Brewer HB, et al. Effect of ACAT inhibition on the progression of coronary atherosclerosis. N Engl J Med 2006;354:1253–1263.
13. Brener SJ, Ivanc TB, Poliszczuk R, et al. Antihypertensive therapy and regression of coronary artery disease: Insights from the Comparison of Amlodipine versus Enalapril to Limit Occurrences of Thrombosis (CAMELOT) and Norvasc for Regression of Manifest Atherosclerotic Lesions by Intravascular Sonographic Evaluation (NORMALISE) trials. Am Heart J 2006;152:1059–1063.
14. Nissen SE, Tuzcu EM, Libby P, et al. Effect of antihypertensive agents on cardiovascular events in patients with coronary disease and normal blood pressure: The CAMELOT study: A randomized controlled trial. JAMA 2004;292:2217–2225.
15. Waters D, Higginson L, Gladstone P, et al. Effects of monotherapy with an HMG-CoA reductase inhibitor on the progression of coronary atherosclerosis as assessed by serial quantitative arteriography. The Canadian Coronary Atherosclerosis Intervention Trial. Circulation 1994;89:959–968.
16. Waters D, Lesperance J, Francetich M, et al. A controlled clinical trial to assess the effect of a calcium channel blocker on the progression of coronary atherosclerosis. Circulation 1990;82:1940–1953.
17. Solymoss BC, Bourassa MG, Campeau L, et al. Effect of increasing metabolic syndrome score on atherosclerotic risk profile and coronary artery disease angiographic severity. Am J Cardiol 2004;93:159–164.
18. Reiber J. Advances in Quantitative Coronary Angiography. Dordrecht, Netherlands: Kluwer Academic Publishers, 1993.
19. Brown BG, Stukovsky KH, Zhao XQ. Simultaneous low-density lipoprotein-C lowering and high-density lipoprotein-C elevation for optimum cardiovascular disease prevention with various drug classes, and their combinations: A meta-analysis of 23 randomized lipid trials. Curr Opin Lipidol 2006;17:631–636.
20. Berry C, L'Allier PL, Gregoire J, et al. Comparison of intravascular ultrasound and quantitative coronary angiography for the assessment of coronary artery disease progression. Circulation 2007;115:1851–1857.
21. Brown BG, Zhao X-Q. Is intravascular ultrasound the gold standard surrogate for clinically relevant atherosclerosis progression? J Am Coll Cardiol 2007;49:933–938.
22. Brown BG. A direct comparison of intravascular ultrasound and quantitative coronary arteriography: Implications for measures of atherosclerosis as clinical surrogates. Circulation 2007;115:1824–1826.
23. Wilson PW, Anderson KM, Castelli WP. Twelve-year incidence of coronary heart disease in middle-aged adults during the era of hypertensive therapy: The Framingham offspring study. Am J Med 1991;90:11–16.
24. MRC/BHF Heart protection study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: A randomised placebo-controlled trial. Lancet 2002;360:7–22.
25. Gordon DJ, Probstfield JL, Garrison RJ, et al. High-density lipoprotein cholesterol and cardiovascular disease. Four prospective American studies. Circulation 1989;79:8–15. 26. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: The Scandinavian Simvast atin Survival Study (4S). Lancet 1994;344:1383–1389.