Low-Density Lipoprotein Cholesterol Level in Patients With Acute Myocardial Infarction Having Percutaneous Coronary Intervention (the Cholesterol Paradox)

Low-Density Lipoprotein Cholesterol Level in Patients With Acute Myocardial Infarction Having Percutaneous Coronary Intervention (the Cholesterol Paradox) Kyung Hoon Cho, MDa, Myung Ho Jeong, MD, PhDa,*, Youngkeun Ahn, MD, PhDa, Young Jo Kim, MD, PhDb, Shung Chull Chae, MD, PhDc, Taek Jong Hong, MD, PhDd, In Whan Seong, MD, PhDe, Jei Keon Chae, MD, PhDf, Chong Jin Kim, MD, PhDg, Myeong Chan Cho, MD, PhDh, Ki Bae Seung, MD, PhDi, Seung Jung Park, MD, PhDj, and other Korea Acute Myocardial Infarction Registry (KAMIR) Investigators The relation between low-density lipoprotein (LDL) cholesterol levels and clinical outcomes after percutaneous coronary intervention (PCI) in patients with acute myocardial infarction (AMI) has not been described. A total of 9,571 eligible patients (mean age 62.6 ⴞ 12.5 years, 6,967 men) who underwent PCI with a final diagnosis of AMI from the Korea Acute Myocardial Infarction Registry (KAMIR) were divided into 5 groups according to LDL cholesterol level: <70, 70 to 99, 100 to 129, 130 to 159, and >160 mg/dl. Clinical outcomes in hospital and 1 and 12 months after PCI in patients with AMI were examined. Age and co-morbidities decreased as LDL cholesterol increased. Patients with higher LDL cholesterol levels had favorable hemodynamic status and laboratory findings. Lifesaving medications, including lipid-lowering drugs, were underused in patients with lower LDL cholesterol levels. Clinical outcomes in hospital and 1 and 12 months after PCI showed better results as LDL cholesterol increased, except for patients with LDL cholesterol levels >160 mg/dl. In a Cox proportional-hazards model, LDL cholesterol level was not an independent predictor of mortality at 12 months, after adjusting for clinical characteristics including demographics and biologic data. In conclusion, the cholesterol paradox in patients with AMI is related to confounding by baseline characteristics associated with survival. More intensive treatment including lipid-lowering therapy for AMI in patients with lower LDL cholesterol level may result in better clinical outcomes. © 2010 Elsevier Inc. All rights reserved. (Am J Cardiol 2010;106:1061–1068) As a part of the Korea Acute Myocardial Infarction Registry (KAMIR) initiative, in this study we investigated the relation between low-density lipoprotein (LDL) cholesterol levels and clinical outcomes after percutaneous coronary intervention (PCI) in patients with acute myocardial infarction (AMI). The aim was to help refine the treatment goal for lipid-lowering therapy and improve outcomes in patients with AMI. Methods KAMIR is a prospective, multicenter, observational data collection registry investigating the risk factors of

a

Chonnam National University Hospital, Gwangju, Korea; bYeungnam University Hospital, Daegu, Korea; cKyungpook National University Hospital, Daegu, Korea; dBusan National University Hospital, Busan, Korea; e Chungnam National University Hospital, Daejon, Korea; fChunbuk National University Hospital, Jeonju, Korea; gKyung Hee University Hospital, Seoul, Korea; hChungbuk National University Hospital, Cheongju, Korea; iCatholic University Hospital, Seoul, Korea; and jAsan Medical Center, Seoul, Korea. Manuscript received March 8, 2010; revised manuscript received and accepted June 2, 2010. This study was performed with the support of the Korean Circulation Society, Seoul, Korea, in commemoration of its 50th anniversary. *Corresponding author: Tel: 82-62-220-6243; fax: 82-62-228-7174. E-mail address: [email protected] (M.H. Jeong). 0002-9149/10/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2010.06.009

mortality in AMI and establishing the universal management for the prevention of AMI, with the support of Korean Circulation Society since November 2005. The database is comprehensive and includes age, gender, body mass index (BMI), vital signs, Killip class, symptom onset time, door-to-needle time, door-to-balloon time, each risk factor, past regular medications, co-morbidities, electrocardiographic locations of myocardial infarction, initial treatment strategy, and drugs. After every PCI procedure, operators filled in standard report forms including angiographic findings, in-hospital complications, and medical therapy in the hospital and completed 12-month follow-up major cardiac events later. From November 2005 to January 2008, 9,571 patients (mean age 62.6 ⫾ 12.5 years, 6,967 men) who underwent PCI with a final diagnosis of AMI were enrolled in the KAMIR. Of these patients, the numbers of patients at 1- and 12-month follow-up were 8,603 (mean age 62.4 ⫾ 12.5, 6,271 men) and 7,485 (mean age 62.5 ⫾ 12.4, 5,454 men), respectively. Blood samples except lipid profiles were collected at admission or before PCI. Overnight fasting blood was drawn for lipid measurements. Lipid profiles were measured directly using routine analyses of local hospitals. Clinical data after discharge were collected from outpatient visits and telephone contacts at 1 and 12 months. www.ajconline.org

1062

The American Journal of Cardiology (www.ajconline.org)

Figure 1. Simple correlation analysis between LDL cholesterol and (A) age (r ⫽ ⫺0.113, p ⬍0.001) and (B) BMI (r ⫽ 0.113, p ⬍0.001).

Table 1 Baseline clinical characteristics according to low-density lipoprotein cholesterol categories Characteristic

Age (years) Men Body mass index (kg/m2) Waist-to-hip ratio Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Heart rate (beats/min) Killip class ⬎1 ST-segment elevation myocardial infarction Family history of coronary artery disease Previous hypertension Previous diabetes mellitus Previous dyslipidemia Smokers Previous coronary artery disease Previous heart failure Previous stroke Previous peripheral vascular disease Left ventricular ejection fraction (%) Serum glucose (mg/dl) Creatinine clearance (ml/min) High-sensitivity C-reactive protein (mg/dl) N-terminal–pro–B-type natriuretic peptide (pg/ml) Total cholesterol (mg/dl) Triglyceride (mg/dl) High-density lipoprotein cholesterol (mg/dl)

Serum Level of LDL Cholesterol (mg/dl) ⬍70 (n ⫽ 840)

70–99 (n ⫽ 2,265)

100–129 (n ⫽ 3,182)

130–159 (n ⫽ 2,075)

ⱖ160 (n ⫽ 1,209)

65 ⫾ 13 609 (73%) 23 ⫾ 4 0.936 ⫾ 0.072 121 ⫾ 31 74 ⫾ 18 78 ⫾ 32 274 (33%) 537 (64%)

64 ⫾ 12 1,735 (77%) 24 ⫾ 3 0.943 ⫾ 0.077 125 ⫾ 29 77 ⫾ 18 75 ⫾ 20 565 (26%) 1,537 (68%)

63 ⫾ 12 2,376 (75%) 24 ⫾ 3 0.944 ⫾ 0.068 129 ⫾ 28 79 ⫾ 17 76 ⫾ 18 656 (22%) 2,079 (65%)

61 ⫾ 12 1,465 (71%) 25 ⫾ 4 0.947 ⫾ 0.083 132 ⫾ 28 81 ⫾ 16 77 ⫾ 23 400 (20%) 1,328 (64%)

60 ⫾ 13 782 (65%) 25 ⫾ 4 0.950 ⫾ 0.088 132 ⫾ 27 81 ⫾ 16 78 ⫾ 18 281 (24%) 771 (64%)

49 (6%)

140 (6%)

221 (7%)

156 (8%)

90 (8%)

453 (54%) 290 (35%) 92 (11%) 466 (56%) 215 (26%) 19 (2.3%) 68 (8%) 12 (1.4%) 50 ⫾ 12 178 ⫾ 88 62 ⫾ 54 16 ⫾ 49

1,122 (50%) 666 (30%) 194 (9%) 1,365 (61%) 406 (18%) 43 (1.9%) 152 (7%) 27 (1.2%) 51 ⫾ 12 170 ⫾ 79 68 ⫾ 33 13 ⫾ 49

1,488 (47%) 805 (25%) 245 (8%) 1,991 (63%) 383 (12%) 30 (0.9%) 193 (6%) 25 (0.8%) 52 ⫾ 12 166 ⫾ 75 72 ⫾ 35 9 ⫾ 42

957 (46%) 504 (24%) 245 (12%) 1,225 (59%) 223 (11%) 17 (0.8%) 111 (5%) 16 (0.8%) 52 ⫾ 11 166 ⫾ 73 75 ⫾ 30 13 ⫾ 59

4,083 ⫾ 8,706

2,499 ⫾ 6,060

1,784 ⫾ 4,438

127 ⫾ 38 121 ⫾ 174 42 ⫾ 25

154 ⫾ 23 120 ⫾ 101 45 ⫾ 25

180 ⫾ 24 125 ⫾ 95 45 ⫾ 14

p Value

Linear p Value*

⬍0.001 ⬍0.001 ⬍0.001 0.005 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 0.040

⬍0.001 0.063

0.290

0.035

525 (44%) 278 (23%) 175 (15%) 687 (57%) 125 (10%) 19 (1.6%) 64 (5%) 6 (0.5%) 52 ⫾ 12 169 ⫾ 80 77 ⫾ 43 16 ⫾ 72

⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 0.001 0.031 0.096 ⬍0.001 ⬍0.001 ⬍0.001 0.002

⬍0.001 ⬍0.001 ⬍0.001 0.530 ⬍0.001 0.009 0.002 0.008

1,701 ⫾ 4,245

1,752 ⫾ 4,479

⬍0.001

209 ⫾ 22 136 ⫾ 102 46 ⫾ 15

253 ⫾ 37 147 ⫾ 93 48 ⫾ 22

⬍0.001 ⬍0.001 ⬍0.001

Data are expressed as mean ⫾ SD or as number (percentage). * Statistical significance for linear by linear association between categorical variables, calculated using the chi-square test for trend.

⬍0.001

Coronary Artery Disease/Cholesterol Paradox in AMI

1063

Table 2 Coronary angiographic and procedural findings according to low-density lipoprotein cholesterol categories Characteristic

Multivessel coronary disease Infarct-related coronary artery Left main Left anterior descending Left circumflex Right Lesion type C Initial TIMI grade 0 flow Stent implantation Stent length (mm) Stent diameter (mm) Stent number Final TIMI grade 3 flow

Serum Level of LDL Cholesterol (mg/dl)

p Value

Linear p Value*

⬍70 (n ⫽ 840)

70–99 (n ⫽ 2,265)

100–129 (n ⫽ 3,182)

130–159 (n ⫽ 2,075)

ⱖ160 (n ⫽ 1,209)

476 (58%)

1,195 (54%)

1,726 (56%)

1,164 (58%)

673 (57%)

0.143

0.253

21 (3%) 357 (44%) 105 (13%) 337 (41%) 363 (49%) 360 (47%) 728 (89%) 25 ⫾ 6 3.2 ⫾ 0.4 1.5 ⫾ 0.8 700 (92%)

41 (2%) 1,020 (46%) 330 (15%) 822 (37%) 992 (49%) 944 (45%) 2,045 (93%) 25 ⫾ 6 3.2 ⫾ 0.4 1.5 ⫾ 0.8 1,969 (94%)

48 (2%) 1,449 (47%) 551 (18%) 1,051 (34%) 1,360 (48%) 1,318 (45%) 2,914 (94%) 25 ⫾ 6 3.2 ⫾ 0.5 1.5 ⫾ 0.8 2,739 (94%)

30 (2%) 997 (49%) 362 (18%) 645 (32%) 904 (48%) 871 (45%) 1,923 (95%) 25 ⫾ 6 3.2 ⫾ 0.4 1.6 ⫾ 0.9 1,819 (94%)

21 (2%) 591 (50%) 205 (17%) 369 (31%) 513 (47%) 498 (44%) 1,119 (95%) 25 ⫾ 6 3.1 ⫾ 0.4 1.5 ⫾ 0.9 1,050 (94%)

0.294 0.019 0.001 ⬍0.001 0.866 0.842 ⬍0.001 0.779 0.404 0.209 0.287

0.154 0.001 0.001 ⬍0.001 0.286 0.361 ⬍0.001

0.341

Data are expressed as number (percentage) or as mean ⫾ SD. * Statistical significance for linear by linear association between categorical variables, calculated using the chi-square test for trend.

Table 3 Treatment strategies for acute myocardial infarction according to low-density lipoprotein cholesterol categories Characteristic

ST-segment elevation myocardial infarction Primary percutaneous coronary intervention Early percutaneous coronary intervention, not primary Elective percutaneous coronary intervention Rescue PCI after thrombolysis Rescue PCI after conservative treatment Non–ST-segment elevation myocardial infarction Early invasive percutaneous coronary intervention Elective percutaneous coronary intervention Rescue PCI after conservative treatment Medical therapy during hospitalization Aspirin Clopidogrel Cilostazol ␤ blockers Angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers Lipid-lowering drugs Medical therapy after discharge Aspirin Clopidogrel Cilostazol ␤ blockers Angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers Lipid-lowering drugs

Serum Level of LDL Cholesterol (mg/dl)

p Value

Linear p Value*

⬍70 (n ⫽ 840)

70–99 (n ⫽ 2,265)

100–129 (n ⫽ 3,182)

130–159 (n ⫽ 2,075)

ⱖ160 (n ⫽ 1,209)

422 (79%) 23 (4%)

1,147 (75%) 96 (6%)

1,531 (74%) 134 (7%)

960 (73%) 85 (6%)

579 (75%) 38 (5%)

0.062 0.231

0.098 0.872

51 (10%) 20 (4%) 19 (4%)

188 (12%) 51 (3%) 52 (3%)

257 (12%) 83 (4%) 70 (3%)

193 (15%) 57 (4%) 29 (2%)

102 (13%) 30 (4%) 20 (3%)

0.045 0.731 0.229

0.013 0.354 0.061

200 (66%) 76 (25%) 25 (8%)

525 (72%) 144 (20%) 57 (8%)

789 (72%) 228 (21%) 79 (7%)

549 (74%) 146 (20%) 50 (7%)

300 (69%) 103 (24%) 33 (8%)

0.114 0.162 0.883

0.563 0.854 0.477

831 (99%) 819 (98%) 246 (29%) 589 (70%) 656 (78%)

2,240 (99%) 2,210 (98%) 721 (32%) 1,626 (72%) 1,825 (81%)

3,151 (99%) 3,122 (99%) 1,055 (33%) 2,369 (75%) 2,575 (81%)

2,055 (99%) 2,046 (99%) 738 (36%) 1,584 (77%) 1,711 (83%)

1,193 (99%) 1,182 (99%) 423 (35%) 922 (77%) 1,028 (86%)

0.983 0.078 0.005 ⬍0.001 ⬍0.001

0.617 0.020 ⬍0.001 ⬍0.001 ⬍0.001

569 (68%)

1,619 (72%)

2,553 (81%)

1,768 (86%)

1,066 (89%)

⬍0.001

⬍0.001

744 (98%) 726 (96%) 236 (31%) 533 (70%) 620 (82%)

2,142 (99%) 2,102 (97%) 739 (34%) 1,526 (70%) 1,771 (82%)

3,015 (99%) 2,952 (97%) 1,074 (35%) 2,241 (73%) 2,459 (81%)

1,963 (98%) 1,941 (97%) 756 (38%) 1,532 (77%) 1,660 (83%)

1,145 (99%) 1,124 (97%) 411 (36%) 889 (77%) 963 (83%)

0.383 0.227 0.010 ⬍0.001 0.084

0.221 0.043 0.004 ⬍0.001 0.090

499 (66%)

1,560 (72%)

2,488 (81%)

1,744 (88%)

1,036 (90%)

⬍0.001

⬍0.001

* Statistical significance for linear by linear association between categorical variables, calculated using the chi-square test for trend.

All patients in this study underwent PCI. The initial treatment strategy for ST-segment elevation myocardial infarction included PCI, thrombolysis, or conservative treatment. For non–ST-segment elevation myocardial infarction, early invasive PCI or conservative treatment was performed

as the initial treatment strategy. The initial treatment strategy for AMI was performed at the discretion of the attending physician on the basis of current guidelines. Primary PCI was defined as emergent PCI performed ⬍12 hours after admission in patients with ST-segment elevation myo-

1064

The American Journal of Cardiology (www.ajconline.org)

Table 4 Clinical outcomes in the hospital period and follow-up at 1 year Variable

Serum Level of LDL Cholesterol (mg/dl)

p Value

Linear p Value*

⬍70

70–99

100–129

130–159

ⱖ160

In-hospital outcomes (n ⫽ 9,571) Success rate of PCI In-hospital complication In-hospital death

(n ⫽ 840) 770 (96%) 168 (20%) 65 (7.7%)

(n ⫽ 2,265) 2,096 (96%) 319 (14%) 61 (2.7%)

(n ⫽ 3,182) 2,951 (97%) 381 (12%) 68 (2.1%)

(n ⫽ 2,075) 1,957 (98%) 183 (9%) 38 (1.8%)

(n ⫽ 1,209) 1,137 (97%) 121 (10%) 29 (2.4%)

0.020 ⬍0.001 ⬍0.001

0.006 ⬍0.001 ⬍0.001

1-month outcomes (n ⫽ 8,603) Composite major cardiac events Death Myocardial infarction Repeat percutaneous coronary intervention Coronary artery bypass grafting

(n ⫽ 748) 88 (12%) 77 (10%) 5 (0.7%) 6 (0.8%)

(n ⫽ 2,019) 112 (6%) 85 (4%) 7 (0.3%) 19 (0.9%)

(n ⫽ 2,867) 142 (5%) 99 (4%) 8 (0.3%) 32 (1.1%)

(n ⫽ 1,880) 77 (4%) 46 (2%) 12 (0.6%) 16 (0.9%)

(n ⫽ 1,089) 59 (5%) 44 (4%) 3 (0.3%) 10 (0.9%)

⬍0.001 ⬍0.001 0.235 0.880

⬍0.001 ⬍0.001 0.858 0.982

3 (0.1%)

3 (0.2%)

2 (0.2%)

0.633

0.113

12-month outcomes (n ⫽ 7,485) Composite major cardiac events Death Myocardial infarction Repeat percutaneous coronary intervention Coronary artery bypass grafting

(n ⫽ 662) 140 (21%) 93 (14%) 7 (1.1%) 36 (5%)

(n ⫽ 1,767) 245 (14%) 116 (7%) 11 (0.6%) 113 (6%)

(n ⫽ 2,490) 345 (14%) 142 (6%) 18 (0.7%) 181 (7%)

(n ⫽ 1,626) 198 (12%) 67 (4%) 14 (0.9%) 109 (7%)

(n ⫽ 940) 136 (15%) 55 (6%) 8 (0.9%) 68 (7%)

⬍0.001 ⬍0.001 0.820 0.468

0.001 ⬍0.001 0.837 0.199

4 (0.6%)

5 (0.3%)

4 (0.2%)

8 (0.5%)

5 (0.5%)

0.208

0.541

0 (0%)

1 (0%)

* Statistical significance for linear by linear association between categorical variables, calculated using the chi-square test for trend.

cardial infarctions. Early invasive PCI was defined as emergent PCI performed ⬍48 hours after admission in patients with non–ST-segment elevation myocardial infarctions. Rescue PCI was defined as emergent PCI for patients with failure of thrombolysis, ongoing ischemic symptoms, cardiogenic shock, severe congestive heart failure, pulmonary edema, or hemodynamically compromising ventricular arrhythmias after thrombolysis or conservative treatment. Optimal evidence-based medical therapies were encouraged in all patients who had no contraindication to drugs before and after discharge. AMI was diagnosed by clinical features, increased biochemical markers, and electrocardiographic findings.1 The left ventricular ejection fraction (LVEF) was determined by echocardiography at admission or before PCI. Estimated creatinine clearance was calculated by the use of the Cockcroft-Gault formula2: creatinine clearance (ml/min) ⫽ [(140 ⫺ age) ⫻ weight (kg)]/[serum creatinine (mg/dl) ⫻ 72], corrected in women by a factor of 0.85. Multivessel disease was defined as ⬎50% diameter stenosis by quantitative coronary angiography in ⱖ2 coronary arteries or a left main coronary artery lesion. Coronary artery lesion type was determined according to American College of Cardiology and American Heart Association classification.3 Successful PCI was defined as a residual stenosis ⬍50% in diameter with final Thrombolysis In Myocardial Infarction (TIMI) grade 3 flow. Patients were categorized into 5 groups according to baseline LDL cholesterol level: ⬍70, 70 to 99, 100 to 129, 130 to 159, and ⱖ160 mg/dl. We analyzed baseline demographic and clinical characteristics, relevant laboratory results, and treatment strategies. In-hospital complications included hypotension, cardiogenic shock, or arrhythmia, requiring cardiopulmonary resuscitation or the insertion of an intra-aortic balloon pump or temporary cardiac pacemaker. Major cardiac events included death, myocardial

infarction, ischemic target vessel revascularization, and coronary artery bypass grafting. Statistical analysis was performed using SPSS version 17.0 (SPSS, Inc., Chicago, Illinois). All continuous variables are reported as mean ⫾ SD. The relation between age and LDL cholesterol was assessed using bivariate correlation analysis. The cumulative incidence of all-cause death was estimated according to the life table method, and Wilcoxon’s rank test was used to evaluate differences between groups. One-way analysis of variance and the chi-square test for independence and trend were used for baseline demographic and clinical comparisons among 5 groups. Univariate analyses using Pearson’s chi-square test or Student’s t test were performed to evaluate the prognostic significance of demographic and clinical variables for 12month mortality. The variables tested in univariate analysis were those known to potentially affect outcomes after myocardial infarction and variables showing variation according to LDL cholesterol quintiles. The variables tested were the following; age, gender, BMI, waist-to-hip ratio, systolic blood pressure, diastolic blood pressure, heart rate, Killip class, ST-segment elevation myocardial infarction, family history of coronary artery disease, smoking history, comorbidities (hypertension, diabetes, dyslipidemia, coronary artery disease, heart failure, stroke, and peripheral vascular disease), the LVEF, glucose, creatinine clearance, highsensitivity C-reactive protein, N-terminal–pro–B-type natriuretic peptide, multivessel disease, infarct-related artery, coronary artery lesion type, initial and final TIMI flow grade, stent implantation, and medical therapy before and after discharge (aspirin, clopidogrel, cilostazol, ␤ blockers, angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers, and lipid-lowering drugs). Cox regression analysis was performed to identify independent predictors of 12-month mortality. All continuous variables were converted into categorical variables. After considering uni-

Coronary Artery Disease/Cholesterol Paradox in AMI

variate analysis results and collinearity among variables, significant variables were entered into a Cox proportionalhazards model in phases. These variables included age, gender, BMI, systolic blood pressure, heart rate, the LVEF, serum glucose, creatinine clearance, high-sensitivity C-reactive protein, N-terminal–pro–B-type natriuretic peptide, ST-segment elevation myocardial infarction, Killip class, family history of coronary artery disease, smoking history, hypertension, diabetes, previous coronary artery disease, previous dyslipidemia, previous heart failure, previous stroke, previous peripheral vascular disease, cilostazol, ␤ blockers, angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers, lipid-lowering drugs, multivessel disease, left main coronary artery disease, lesion type C, initial TIMI flow grade 0, final TIMI flow grade 3, and stent implantation. All LDL cholesterol groups were independently entered into the model, with the group with LDL cholesterol 130 to 159 mg/dl being the reference. The enter method was used to select the predictive variables. To examine the collinearity among the independent variables, eigenvalues, condition index, and variance proportions were checked using linear regression analysis, which showed no significant collinearity among variables. Results The mean LDL cholesterol level in all patients was 118 ⫾ 43 mg/dl. The mean LDL cholesterol levels were 116 ⫾ 40 mg/dl in men and 123 ⫾ 49 mg/dl in women (p ⬍0.001). LDL cholesterol levels were correlated weakly with age and decreased with increasing age. The correlation between LDL cholesterol level and BMI was positive but weak (Figure 1). Baseline clinical characteristics and laboratory findings of the patients’ groups are listed in Table 1. A history of hypertension, diabetes, previous coronary artery disease, previous heart failure, previous stroke, and previous peripheral vascular disease decreased as LDL cholesterol increased. A U-shaped relation between LDL cholesterol level and a history of previous dyslipidemia was observed. The use of lipid-lowering therapy before admission decreased as LDL cholesterol increased, from 9.3% for patients with LDL cholesterol ⬍70 mg/dl to 4.2% for those with LDL cholesterol ⱖ160 mg/dl (linear p ⬍0.001, data not shown). Waist-to-hip ratio and creatinine clearance decreased as LDL cholesterol level decreased. Patients with higher LDL cholesterol levels had higher systolic blood pressures and LVEFs and lower Killip classes, serum glucose levels, and N-terminal–pro– B-type natriuretic peptide levels, except for patients with LDL cholesterol levels ⱖ160 mg/dl. Table 2 lists the findings of coronary angiography and intervention. Proportions of lesions of the left anterior descending coronary artery and stent implantation increased, while that of right coronary artery lesions decreased, as LDL cholesterol level increased. An increasing trend of the proportion of left circumflex coronary artery lesions was observed with increasing values of LDL cholesterol, except for patients with LDL cholesterol levels ⱖ160 mg/dl. There were no significant differences in the proportions of mul-

1065

Figure 2. Cumulative survival rate according to LDL cholesterol categories (life table method).

tivessel disease, lesion type C, initial TIMI grade 0 flow, and final TIMI grade 3 flow. Table 3 lists the types of PCI and medical therapies. There were no significant differences among groups in performing the initial treatment strategy and early or rescue PCI. The use of cilostazol, ␤ blockers, angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers, and lipid-lowering drugs during hospitalization and cilostazol, ␤ blockers, and lipid-lowering drugs after discharge increased as LDL cholesterol level increased, with significant differences among groups. Clinical outcomes in hospital and 1 and 12 months after PCI are listed in Table 4. Of the total of 9,571 patients, the success rate of PCI was 97% (n ⫽ 8,911). The in-hospital complication and in-hospital mortality rates were 12% (n ⫽ 1,172) and 3% (n ⫽ 261), respectively. The 1-month composite major cardiac event rate was 6% (n ⫽ 478), including 351 deaths (4%), 35 recurrent myocardial infarctions (0.4%), 83 repeat PCIs (1%), and 9 coronary artery bypass grafting procedures (0.1%). The 12-month composite major cardiac event rate was 14% (n ⫽ 1,064), including 473 deaths (6%), 58 recurrent myocardial infarctions (0.8%), 507 repeat PCIs (7%), and 26 coronary artery bypass grafting procedures (0.3%). The success rate of PCI, in-hospital complications, and in-hospital mortality revealed better results as LDL cholesterol increased, except for patients with LDL cholesterol levels ⱖ160 mg/dl. The composite major cardiac event and mortality rates at 1 and 12 months decreased as LDL cholesterol increased, except for patients with LDL cholesterol levels ⱖ160 mg/dl. The 12-month mortality rate was significantly higher in patients with LDL cholesterol levels ⬍70 mg/dl than in those with LDL cholesterol levels ⬎70 mg/dl (Figure 2, Table 4). We performed a Cox regression analysis to identify the prognostic value of LDL cholesterol level for 1-year mortality (Table 5). With an LDL cholesterol level

1066

The American Journal of Cardiology (www.ajconline.org)

Table 5 Prognostic value of low-density lipoprotein cholesterol level for 1-year mortality (Cox proportional-hazards model) LDL Cholesterol Level ⬍70 70–99 100–129 130–159§ ⱖ160

Hazard Ratio (95% Confidence Interval) No Adjustment

p Value

Model 1*

p Value

Model 2†

p Value

Model 3‡

p Value

3.62 (2.64–4.96) 1.61 (1.19–2.17) 1.39 (1.04–1.86) 1.0 1.43 (1.00–2.05)

⬍0.001 0.002 0.025

2.81 (2.05–3.86) 1.42 (1.05–1.93) 1.27 (0.95–1.71) 1.0 1.53 (1.07–2.18)

⬍0.001 0.022 0.104

1.49 (0.85–2.60) 1.05 (0.64–1.74) 1.21 (0.76–1.93) 1.0 1.17 (0.63–2.19)

0.166 0.836 0.419

1.46 (0.81–2.65) 0.972 (0.57–1.65) 1.27 (0.78–2.07) 1.0 1.37 (0.72–2.60)

0.212 0.915 0.333

0.048

0.020

0.616

0.344

* Adjusted for age and gender. † Adjusted for BMI, systolic blood pressure, heart rate, the LVEF, serum glucose, creatinine clearance, high-sensitivity C-reactive protein, N-terminal– pro–B-type natriuretic peptide, ST-segment elevation myocardial infarction, Killip class ⬎1, family history of coronary artery disease, smoking history, and co-morbidities (hypertension, diabetes, previous coronary artery disease, previous dyslipidemia, previous heart failure, previous stroke, previous peripheral vascular disease) with variables in model 1. ‡ Adjusted for medical therapy before and after discharge (cilostazol, ␤ blockers, angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers, and lipid-lowering drugs) and angiographic findings (multivessel disease, left main coronary artery disease, lesion type C, initial TIMI flow grade 0, final TIMI flow grade 3, and stent implantation) with variables in model 2. § Reference group.

Figure 3. Independent predictors of 12-month mortality in the fully adjusted model.

of 130 to 159 mg/dl as the reference group, the risk for 1-year mortality was higher for LDL cholesterol levels of ⱖ160, 100 to 129, 70 to 99, and ⬍70 mg/dl in the unadjusted model. In adjusted model 1, LDL cholesterol levels of 70 to 99 and ⬍70 mg/dl were mortality risk predictors. However, LDL cholesterol level was not an independent predictor of mortality at 12 months in adjusted model 2 and fully adjusted model 3. Independent predictors of 12-month mortality after PCI in patients with AMI are shown in Figure 3. The use of lipidlowering drugs had a borderline significant prognostic value in the prediction of 12-month mortality (hazard ratio 0.70, 95% confidence interval 0.48 to 1.00, p ⫽ 0.05; data not shown). Discussion The present study demonstrated that higher LDL cholesterol levels, except for patients with LDL cholesterol levels ⱖ160 mg/dl, were related to better clinical outcomes after PCI in patients with AMI. However, this cholesterol paradox may be due to several confounders. Of the independent predictive factors of 12-month mortality after PCI in patients with AMI, age, systolic blood pressure, Killip class, the LVEF, creatinine clearance, N-terminal–pro–B-type na-

triuretic peptide level, and the use of angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers may play an important role in significant results in the clinical outcomes of each LDL cholesterol group. Hypercholesterolemia is a well-known risk factor for the development of coronary artery disease.4,5 Statins are highly effective in reducing LDL cholesterol levels by reducing the endogenous synthesis of cholesterol. Also, statins have many effects beyond simply lowering cholesterol (so-called pleiotropic effects).6 Statins have been shown to reduce mortality and morbidity in several randomized controlled trials in patients with coronary artery disease.7–11 However, Rauchhaus et al12 demonstrated that higher cholesterol levels were associated with better survival in patients with chronic heart failure. The mechanisms for this cholesterol paradox are not clear. They suggested that circulating lipoproteins have a potentially beneficial role by neutralizing lipopolysaccharide, which is a possible source of the immunologic activation in chronic heart failure.12 Witte and Clark13 suggested that low cholesterol levels may reflect reduced food intake and reduced intestinal absorption due to bowel edema and may possibly be a result of increased metabolic stress.

Coronary Artery Disease/Cholesterol Paradox in AMI

The present study showed that in patients with AMI, clinical outcomes in hospital and at 1 and 12 months decreased as LDL cholesterol increased, except for patients with LDL cholesterol levels ⱖ160 mg/dl. And survival rates were markedly lower in patients with LDL cholesterol levels ⬍70 mg/dl than in those with LDL cholesterol levels ⱖ70 mg/dl. Especially in-hospital mortality in the former group was relatively high. Overall, patients with AMI with lower LDL cholesterol levels were older and had more co-morbidities and unfavorable hemodynamic status. Especially the differences among groups were more apparent in patients with LDL cholesterol levels ⬍70 mg/dl. Higher LDL cholesterol levels may reflect better nutritional and health status, which are likely related to better tolerance of acute medical stress. In the Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes With Early Implementation of the ACC/AHA Guidelines (CRUSADE) study, patients with non–ST-segment elevation acute coronary syndromes with histories of hypercholesterolemia had a lower risk for in-hospital morality after adjustment for clinical variables plus previous statin use (hazard ratio 0.74, 95% confidence interval 0.68 to 0.80).14 This study suggested that previously documented hypercholesterolemia may be a general marker for patients who have had more contact with the medical field before admission.14 However, in our study, the use of lipid-lowering drugs before admission decreased as LDL cholesterol increased. This suggests that a history of previous medical contact does not exert an influence on the cholesterol paradox in patients with AMI. It is well known that smokers have better survival after AMI, simply because they are younger and have more thrombus and less atherosclerosis (the smoker paradox).15 Also, several studies have previously reported that the higher the BMI, the better the clinical outcome after PCI (the obesity paradox).16,17 The mechanisms of this obesity paradox are not clear. However, recent studies have demonstrated that the obesity paradox may be related to confounding by baseline characteristics associated with survival.18–20 Similarly, in our study, Cox regression analysis revealed that LDL cholesterol level was not an independent predictor of mortality at 12 months after adjustment for clinical variables. Hence, the cholesterol paradox in patients with AMI may be associated with several confounders. Of the independent predictive factors of 12-month mortality after PCI in patients with AMI, age, systolic blood pressure, Killip class, the LVEF, creatinine clearance, and N-terminal–pro– B-type natriuretic peptide level may be included among those confounders. Also, we observed that lifesaving medications including lipid-lowering drugs were underused in patients with lower LDL cholesterol levels. This may be attributable to an unfavorable hemodynamic status such as a blood pressure, Killip class, and the LVEF and laboratory results such as creatinine clearance and lipid profiles, with decreasing values of LDL cholesterol. These differences on medical therapy between groups may contribute to the worse prognosis in patients with lower LDL cholesterol levels. In the present study, the use of angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers was an independent predictor of favorable 12-month clinical outcomes after PCI

1067

in patients with AMI. Also, the use of lipid-lowering drugs had borderline significant prognostic value for favorable 12-month clinical outcomes. More intensive treatment including lipid-lowering therapy for AMI in patients with lower LDL cholesterol levels may result in better clinical outcomes. There were some limitations to this analysis. First, in most cases, blood samples for lipid profile were taken at least several hours (“overnight fasting blood”) after the onset of infarction. Therefore, the LDL cholesterol levels may be influenced by the infarction. However, a recent study showed that LDL cholesterol levels changed relatively little over 4 days after acute coronary syndromes.21 Second, a single value of LDL cholesterol at admission was used, so we could not assess the effect of the change in LDL cholesterol level on clinical outcomes during follow-up. We focused on the clinical significance of initial LDL cholesterol levels on clinical outcomes in patients with AMI. Third, our patients were analyzed in the short term (1 year), and the follow-up rate was relatively low (7,485 of 9,571 [78.2%]). Patients who were lost to follow-up did not visit the outpatient clinic and could not be contacted by phone. Nevertheless, the number of follow-up patients at 12 months was substantial (n ⫽ 7,485). Acknowledgment: The KAMIR investigators were Myung Ho Jeong, MD, Youngkeun Ahn, MD, Shung Chull Chae, MD, Jong Hyun Kim, MD, Seung Ho Hur, MD, Young Jo Kim, MD, In Whan Seong, MD, Dong Hoon Choi, MD, Jei Keon Chae, MD, Taek Jong Hong, MD, Jae Young Rhew, MD, Doo Il Kim, MD, In Ho Chae, MD, Jung Han Yoon, MD, Bon Kwon Koo, MD, Byung Ok Kim, MD, Myoung Yong Lee, MD, Kee Sik Kim, MD, Jin Yong Hwang, MD, Myeong Chan Cho, MD, Seok Kyu Oh, MD, Nae Hee Lee, MD, Kyoung Tae Jeong, MD, Seung Jea Tahk, MD, Jang Ho Bae, MD, Seung Woon Rha, MD, Keum Soo Park, MD, Chong Jin Kim, MD, Kyoo Rok Han, MD, Tae Hoon Ahn, MD, Moo Hyun Kim, MD, Ki Bae Seung, MD, Wook Sung Chung, MD, Ju Young Yang, MD, Chong Yun Rhim, MD, Hyeon Cheol Gwon, MD, Seong Wook Park, MD, Young Youp Koh, MD, Seung Jae Joo, MD, Soo Joong Kim, MD, Dong Kyu Jin, MD, Jin Man Cho, MD, Yang Soo Jang, MD, Jeong Gwan Cho, MD, and Seung Jung Park, MD. 1. French JK, White HD. Clinical implications of the new definition of myocardial infarction. Heart 2004;90:99 –106. 2. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976;16:31– 41. 3. Gibson CM, Schomig A. Coronary and myocardial angiography: angiographic assessment of both epicardial and myocardial perfusion. Circulation 2004;109:3096 –3105. 4. Martin MJ, Hulley SB, Browner WS, Kuller LH, Wentworth D. Serum cholesterol, blood pressure, and mortality: implications from a cohort of 361,662 men. Lancet 1986;2:933–936. 5. Pekkanen J, Linn S, Heiss G, Suchindran CM, Leon A, Rifkind BM, Tyroler HA. Ten-year mortality from cardiovascular disease in relation to cholesterol level among men with and without preexisting cardiovascular disease. N Engl J Med 1990;322:1700 –1707. 6. Tousoulis D, Charakida M, Stefanadi E, Siasos G, Latsios G, Stefanadis C. Statins in heart failure. Beyond the lipid lowering effect. Int J Cardiol 2007;115:144 –150.

1068

The American Journal of Cardiology (www.ajconline.org)

7. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 1994;344:1383–1389. 8. Sacks FM, Pfeffer MA, Moye LA, Rouleau JL, Rutherford JD, Cole TG, Brown L, Warnica JW, Arnold JM, Wun CC, Davis BR, Braunwald E; Cholesterol and Recurrent Events Trial Investigators. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. N Engl J Med 1996;335: 1001–1009. 9. The Long-Term Intervention With Pravastatin in Ischaemic Disease (LIPID) Study Group. Prevention of cardiovascular events and death with pravastatin in patients with coronary heart disease and a broad range of initial cholesterol levels. N Engl J Med 1998;339:1349 –1357. 10. Schwartz GG, Olsson AG, Ezekowitz MD, Ganz P, Oliver MF, Waters D, Zeiher A, Chaitman BR, Leslie S, Stern T. Effects of atorvastatin on early recurrent ischemic events in acute coronary syndromes: the MIRACL study: a randomized controlled trial. JAMA 2001;285: 1711–1718. 11. Ray KK, Cannon CP, McCabe CH, Cairns R, Tonkin AM, Sacks FM, Jackson G, Braunwald E. Early and late benefits of high-dose atorvastatin in patients with acute coronary syndromes: results from the PROVE IT-TIMI 22 trial. J Am Coll Cardiol 2005;46:1405–1410. 12. Rauchhaus M, Clark AL, Doehner W, Davos C, Bolger A, Sharma R, Coats AJ, Anker SD. The relationship between cholesterol and survival in patients with chronic heart failure. J Am Coll Cardiol 2003; 42:1933–1940. 13. Witte KK, Clark AL. Nutritional abnormalities contributing to cachexia in chronic illness. Int J Cardiol 2002;85:23–31. 14. Wang TY, Newby LK, Chen AY, Mulgund J, Roe MT, Sonel AF, Bhatt DL, DeLong ER, Ohman EM, Gibler WB, Peterson ED. Hypercholesterolemia paradox in relation to mortality in acute coronary syndrome. Clin Cardiol 2009;32:E22–E28. 15. Anderson JL, Adams CD, Antman EM, Bridges CR, Califf RM, Casey DE Jr, Chavey WE II, Fesmire FM, Hochman JS, Levin TN, Lincoff AM, Peterson ED, Theroux P, Wenger NK, Wright RS, Smith SC Jr, Jacobs AK, Halperin JL, Hunt SA, Krumholz HM, Kushner FG, Lytle BW, Nishimura R, Ornato JP, Page RL, Riegel B. ACC/AHA 2007

16.

17.

18.

19.

20.

21.

guidelines for the management of patients with unstable angina/non-ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines for the Management of Patients With Unstable Angina/Non-STElevation Myocardial Infarction) developed in collaboration with the American College of Emergency Physicians, the Society for Cardiovascular Angiography and Interventions, and the Society of Thoracic Surgeons endorsed by the American Association of Cardiovascular and Pulmonary Rehabilitation and the Society for Academic Emergency Medicine. J Am Coll Cardiol 2007;50:e1– e157. Gruberg L, Weissman NJ, Waksman R, Fuchs S, Deible R, Pinnow EE, Ahmed LM, Kent KM, Pichard AD, Suddath WO, Satler LF, Lindsay J Jr. The impact of obesity on the short-term and long-term outcomes after percutaneous coronary intervention: the obesity paradox? J Am Coll Cardiol 2002;39:578 –584. Mehta L, Devlin W, McCullough PA, O’Neill WW, Skelding KA, Stone GW, Boura JA, Grines CL. Impact of body mass index on outcomes after percutaneous coronary intervention in patients with acute myocardial infarction. Am J Cardiol 2007;99:906 –910. Nikolsky E, Stone GW, Grines CL, Cox DA, Garcia E, Tcheng JE, Griffin JJ, Guagliumi G, Stuckey T, Turco M, Negoita M, Lansky AJ, Mehran R. Impact of body mass index on outcomes after primary angioplasty in acute myocardial infarction. Am Heart J 2006;151:168 – 175. Zeller M, Steg PG, Ravisy J, Lorgis L, Laurent Y, Sicard P, JaninManificat L, Beer JC, Makki H, Lagrost AC, Rochette L, Cottin Y. Relation between body mass index, waist circumference, and death after acute myocardial infarction. Circulation 2008;118:482– 490. Kang WY, Jeong MH, Ahn YK, Kim JH, Chae SC, Kim YJ, Hur SH, Seong IW, Hong TJ, Choi DH, Cho MC, Kim CJ, Seung KB, Chung WS, Jang YS, Rha SW, Bae JH, Cho JG, Park SJ. Obesity paradox in Korean patients undergoing primary percutaneous coronary intervention in ST-segment elevation myocardial infarction. J Cardiol 2010; 55:84 –91. Pitt B, Loscalzo J, Ycas J, Raichlen JS. Lipid levels after acute coronary syndromes. J Am Coll Cardiol 2008;51:1440 –1445.