Abstract
Background
The comparative prognostic value of low density lipoprotein-cholesterol (LDL-C) and C-reactive protein (CRP) in patients with stable coronary artery disease (CAD) treated with percutaneous coronary intervention (PCI) and statin therapy is poorly investigated.
Methods
The study included 7595 patients with stable CAD treated with PCI. Based on a cut-off of 100 mg/dl for LDL-C and 3 mg/L for CRP, patients were divided into 4 groups: patients with LDL-C ≤ 100 mg/dl and CRP ≤ 3 mg/L (n = 2795); patients with LDL-C > 100 mg/dl and CRP ≤ 3 mg/L (n = 2091); patients with LDL-C ≤ 100 mg/dl and CRP > 3 mg/L (n = 1296); and patients with LDL-C > 100 mg/dl and CRP > 3 mg/L (n = 1413). Statins at discharge were prescribed in all patients. The primary outcome was 1-year all-cause mortality.
Results
One-year mortality was 2.1% (160 deaths): 1.2% (33 deaths) among patients with LDL-C ≤ 100 mg/dl and CRP ≤ 3 mg/L, 1.4% (28 deaths) among patients with LDL-C > 100 mg/dl and CRP ≤ 3 mg/L, 4.8% (60 deaths) among patients with LDL-C ≤ 100 mg/dl and CRP > 3 mg/L and 2.9% (39 deaths) among patients with LDL-C > 100 mg/dl and CRP > 3 mg/L (P < 0.001). After adjustment, CRP (hazard ratio [HR] = 1.64, 95% confidence interval [CI] 1.33–2.02, for 1 standard deviation increase in the logarithmic scale) but not LDL-C (HR = 1.03 [0.90–1.17], for 30 mg/dl increase) showed an independent association with 1-year mortality. CRP (P = 0.045) but not LDL-C (P = 0.294) increased the discriminatory power of multivariable model for prediction of mortality.
Conclusion
In patients with stable CAD treated with PCI and statin therapy, CRP but not LDL-C was independently associated with increased risk of 1-year mortality.
1
Introduction
Low density lipoprotein-cholesterol (LDL-C) and C-reactive protein (CRP) are among the most commonly measured biomarkers in clinical medicine. LDL-C is a well-established risk factor for coronary artery disease (CAD) and a consistent associate of long term morbidity and mortality in patients with CAD . The high efficacy of statins in reducing cardiovascular events is mediated to a great extent, although not exclusively, by their ability to reduce LDL-C . A recent meta-analysis of randomized trials showed that statin therapy markedly reduced LDL-C and mortality, being associated with a 12% proportional reduction in all-cause mortality per mmol/L reduction in LDL-C level . Although recent studies have shown that LDL-C is decreasing in US population , LDL-C still remains a major target of lipid-lowering therapy in patients with CAD. CRP is well-known marker of inflammation that has shown a strong association with morbidity and mortality related to CAD . Several studies have demonstrated that high CRP levels are associated with mortality after percutaneous coronary intervention (PCI) . Statins through their pleiotropic effects reduce the CRP level , which appears to be associated with clinical benefit in patients with CAD . Some studies suggested that serum cholesterol and CRP have equivalent prognostic utility . Nevertheless, it has recently been demonstrated that high CRP levels continue to predict an increased risk of mortality after PCI irrespective LDL-C levels, even in LDL-C levels as low as < 70 mg/dl or that LDL-C has lost its association with mortality after PCI . Apart from LDL-C and CRP lowering effects, statins also reduce the odds for presentation with an acute myocardial infarction . Thus, in the current era of broad use of statins in the setting of secondary prevention in patients with CAD after PCI, the comparative prognostic value of LDL-C and CRP remains poorly investigated. We undertook this study to investigate the association between LDL-C and CRP with clinical outcome and to compare the prognostic value of these biomarkers in patients with stable CAD treated with PCI and chronic statin therapy.
2
Methods
2.1
Patients
The study included 7595 consecutive patients with stable CAD who were treated at the Deutsches Herzzentrum in Munich between March 2000 and December 2009 in the setting of clinical trials designed to assess efficacy of antithrombotic regimens in patients undergoing PCI. Blood samples were obtained in all patients with the intention to investigate prognostic values of various biomarkers. Although, data are prospectively collected, including the follow-up information, by design, the present study represents a retrospective analysis. To be included in the study, patients had to have been treated with PCI for stable CAD, discharged on statin therapy and had LDL-C and CRP data available. Patients with acute coronary syndromes, advanced kidney disease (serum creatinine > 2 mg/dl), acute infections, or known malignant diseases with expected life expectancy < 1 year were excluded. All patients gave written informed consent for angiographic examination, PCI and blood sampling. The study conforms to the Declaration of Helsinki and was approved by the institutional ethics committee.
2.2
Definitions
Clinically, stable CAD was defined in case of typical chest pain that did not change its pattern within the last 2 months. Angiographic diagnosis of CAD required documentation of coronary stenosis with ≥ 50% lumen obstruction in, at least, one of the three major coronary arteries. Cardiovascular risk factors were defined as follows: arterial hypertension was defined as a systolic blood pressure of more than 140 mmHg and/or diastolic blood pressure of more than 90 mmHg on at least two separate occasions or in the presence of active treatment with antihypertensive agents; hypercholesterolemia was defined as a total cholesterol ≥ 220 mg/dl or if patients were under active treatment with lipid-lowering therapy; diabetes was defined if patients were under treatment with insulin or hypoglycemic drugs or otherwise as a documented fasting blood glucose test (glucose ≥ 126 mg/dl), a blood glucose ≥ 200 mg/dl any time or an abnormal 2-h post-load glucose test (post-load plasma glucose ≥ 200 mg/dl), according to the World Health Organization criteria; current smoking was defined as regular smoking in the prior 6 months. Patient weight and height were measured during the hospital course and the body mass index was calculated. The creatinine clearance was estimated using the Cockcroft–Gault equation .
Stent implantation and peri-PCI care were performed as per standard practice. Before PCI procedure, patients received aspirin (325–500 mg) and a loading dose of clopidogrel (600 mg). Post-PCI antithrombotic therapy included aspirin 200 mg/day continuously and clopidogrel (75 mg/day) for ≥ 1 month if bare-metal stents were implanted or ≥ 6 months if drug-eluting stents were used. At discharge, statins were prescribed to all patients. Other medications were left at the discretion of the attending physician.
2.3
Biochemical measurements
Blood samples were obtained before angiography. Blood was collected into tubes containing lithium–heparin as anticoagulant. Cholesterol and triglycerides were determined in the lithium–heparin plasma samples using standard enzymatic, colorimetric methods. The measuring range was 3.87–800 mg/dl (0.1–20.7 mmol/L) for cholesterol and 8.85–885 mg/dl (0.1–10.0 mmol/L) for triglycerides. LDL-C and HDL-C were measured using homogeneous enzymatic colorimetric assays. The measuring range was 3.86–548 mg/dl (0.10–14.2 mmol/L) for LDL-C and 3–120 mg/dl (0.08–3.10 mmol/L) for HDL-C. In patients with triglyceride concentrations < 400 mg/dl, LDL-C was calculated according to the Friedewald’s formula. All assays were run on a Cobas Integra 800 analyzer (Roche Diagnostics). Plasma concentrations of C-reactive protein were measured using a fully automated, high sensitivity, latex-enhanced immunoturbidometric assay on a Cobas Integra (Roche Diagnostics, Mannheim, Germany). The C-reactive protein assay has a high analytical sensitivity of 0.085 mg/L and a measuring range up to 160 mg/L. Creatinine was measured using a kinetic colorimetric assay based on the compensated Jaffe method. Laboratory personnel involved in the biochemical measurements were unaware of clinical or follow-up data of the patients.
2.4
Outcomes and follow-up
The primary outcome was all-cause 1-year mortality. Secondary outcomes included: cardiac mortality, non-cardiac mortality, non-fatal myocardial infarction, stroke and stent thrombosis. Information on deaths was obtained from hospital records, death certificates or phone contact with the referring physician(s), relatives of the patient, insurance companies and registration of address office. Cardiac death was defined according to the Academic Research Consortium criteria . Non-fatal myocardial infarction was diagnosed in case of development of new abnormal Q waves in ≥ 2 contiguous precordial leads or ≥ 2 adjacent limb leads; or an elevation of creatine kinase-myocardial band (CK-MB) > 2 times (or > 3 times for the 48 h after PCI) the upper limit of normal. The diagnosis of stroke required the development of acute neurologic deficits that were confirmed by computed tomography or magnetic resonance imaging of the head. Definite stent thrombosis was defined using the Academic Research Consortium criteria . The follow-up protocol included a telephone interview at 1 month, a clinic visit at 6 months and a telephone interview at 12 months after the index event. In addition, if patients had cardiac complaints any time during the follow-up a complete clinical, electrocardiographic, and laboratory evaluation was done. Follow-up data were collected by personnel unaware of CRP or plasma lipid levels.
2.5
Statistical analysis
Data are presented as median (25th to 75th percentiles) or counts and proportions (%). One-sample Kolmogorov–Smirnov test was used to assess the normality of data distribution. Categorical data were compared with chi-square test. Continuous data were compared with the Kruskal–Wallis rank-sum test. Survival analysis was performed by applying the Kaplan–Meier method and log-rank test. Multivariable Cox proportional hazards model was used to assess the association between LDL-C and/or CRP with the outcome while adjusting for other potential confounders. All parameters of Table 1 (with the exception of serum creatinine) were entered into the model. CRP was entered after logarithmic transformation and the risk estimates (in this case, hazard ratio) were calculated per standard deviation increase in the CRP logarithmic scale. The discriminatory power of the multivariable models regarding mortality was assessed by calculating the Harrell’s c statistic before and after inclusion of LDL-C, CRP or both and by calculating the integrated discrimination improvement (IDI) . Comparison between c statistics of the models before and after inclusion of LDL-C, CRP, or both alongside other baseline variables was enabled by the Somer’s D package. All analyses were performed using the R statistical package. A 2-tailed P < 0.05 was considered to indicate statistical significance.
| Characteristic | LDL-C ≤ 100 mg/dl and CRP ≤ 3 mg/L (n = 2795) | LDL-C > 100 mg/dl and CRP ≤ 3 mg/L (n = 2091) | LDL-C ≤ 100 mg/dl and CRP > 3 mg/L (n = 1296) | LDL-C > 100 mg/dl and CRP > 3 mg/L (n = 1413) | P value |
|---|---|---|---|---|---|
| Age (years) | 67.3 [60.0; 73.6] | 65.9 [58.7; 73.0] | 69.0 [61.1; 75.6] | 66.7 [58.9; 73.0] | < 0.001 |
| Women | 485 (17.4) | 447 (21.4) | 285 (21.0) | 412 (29.2) | < 0.001 |
| Body mass index (kg/m 2 ) | 26.7 [24.4; 29.1] | 26.7 [24.4; 29.1] | 27.7 [25.2; 30.8] | 27.8 [25.2; 31.0] | < 0.001 |
| Diabetes | 803 (28.7) | 450 (21.5) | 453 (35.0) | 349 (24.7) | < 0.001 |
| Insulin requiring | 238 (8.5) | 121 (5.8) | 163 (12.6) | 113 (8.0) | < 0.001 |
| Arterial hypertension | 2042 (73.1) | 1514 (72.4) | 951 (73.4) | 1038 (73.5) | 0.890 |
| Current smoking | 239 (8.6) | 246 (11.8) | 180 (13.9) | 266 (18.8) | < 0.001 |
| History of hypercholesterolemia | 2230 (79.8) | 1567 (74.9) | 976 (75.3) | 1000 (70.8) | < 0.001 |
| Previous myocardial infarction | 1123 (40.2) | 644 (30.8) | 538 (41.5) | 437 (30.9) | < 0.001 |
| Previous CABG surgery | 479 (17.1) | 282 (13.5) | 263 (20.3) | 224 (15.9) | < 0.001 |
| Serum creatinine (mg/dl) | 0.96 [0.80-1.10] | 0.94 [0.80-1.10] | 1.00 [0.88-1.25] | 1.00 [0.87-1.20] | < 0.001 |
| Creatinine clearance (ml/min) | 83.2 [65.4; 103.1] | 84.2 [66.4; 104.4] | 76.5 [57.1; 101.2] | 79.3 [60.4; 104.7] | < 0.001 |
| CRP (mg/L) | 1.0 [0.4; 1.7] | 1.2 [0.6; 1.9] | 6.2 [4.1; 11.0] | 6.7 [4.5; 12.0] | < 0.001 |
| LDL-C (mg/dl) | 78.2 [65.5; 88.9] | 124.0 [111.0; 146.0] | 79.0 [65.6; 89.9] | 129.0 [114.0; 155.0] | < 0.001 |
| HDL-C (mg/dl) | 49.6 [41.2; 60.4] | 51.3 [43.4; 61.0] | 44.4 [36.6; 54.6] | 46.8 [39.4; 56.1] | < 0.001 |
| Triglycerides (mg/dl) | 112.0 [79.7; 162.0] | 130.0 [94.5; 182.0] | 120.0 [88.0; 176.0] | 140.0 [103.0; 189.0] | < 0.001 |
| Nr. of narrowed coronary arteries | < 0.001 | ||||
| One-vessel disease | 339 (12.1) | 407 (19.5) | 142 (11.0) | 280 (19.8) | |
| Two-vessel disease | 750 (26.8) | 562 (26.9) | 319 (24.6) | 395 (28.0) | |
| Three-vessel disease | 1706 (61.1) | 1122 (53.6) | 835 (64.4) | 738 (52.2) | |
| Multivessel disease | 2456 (87.9) | 1684 (80.5) | 1154 (89.0) | 1133 (80.2) | < 0.001 |
| Left ventricular ejection fraction (%) | 59.0 [50.0; 63.0] | 60.0 [52.0; 65.0] | 56.0 [46.0; 62.0] | 58.0 [48.0; 64.0] | < 0.001 |
2
Methods
2.1
Patients
The study included 7595 consecutive patients with stable CAD who were treated at the Deutsches Herzzentrum in Munich between March 2000 and December 2009 in the setting of clinical trials designed to assess efficacy of antithrombotic regimens in patients undergoing PCI. Blood samples were obtained in all patients with the intention to investigate prognostic values of various biomarkers. Although, data are prospectively collected, including the follow-up information, by design, the present study represents a retrospective analysis. To be included in the study, patients had to have been treated with PCI for stable CAD, discharged on statin therapy and had LDL-C and CRP data available. Patients with acute coronary syndromes, advanced kidney disease (serum creatinine > 2 mg/dl), acute infections, or known malignant diseases with expected life expectancy < 1 year were excluded. All patients gave written informed consent for angiographic examination, PCI and blood sampling. The study conforms to the Declaration of Helsinki and was approved by the institutional ethics committee.
2.2
Definitions
Clinically, stable CAD was defined in case of typical chest pain that did not change its pattern within the last 2 months. Angiographic diagnosis of CAD required documentation of coronary stenosis with ≥ 50% lumen obstruction in, at least, one of the three major coronary arteries. Cardiovascular risk factors were defined as follows: arterial hypertension was defined as a systolic blood pressure of more than 140 mmHg and/or diastolic blood pressure of more than 90 mmHg on at least two separate occasions or in the presence of active treatment with antihypertensive agents; hypercholesterolemia was defined as a total cholesterol ≥ 220 mg/dl or if patients were under active treatment with lipid-lowering therapy; diabetes was defined if patients were under treatment with insulin or hypoglycemic drugs or otherwise as a documented fasting blood glucose test (glucose ≥ 126 mg/dl), a blood glucose ≥ 200 mg/dl any time or an abnormal 2-h post-load glucose test (post-load plasma glucose ≥ 200 mg/dl), according to the World Health Organization criteria; current smoking was defined as regular smoking in the prior 6 months. Patient weight and height were measured during the hospital course and the body mass index was calculated. The creatinine clearance was estimated using the Cockcroft–Gault equation .
Stent implantation and peri-PCI care were performed as per standard practice. Before PCI procedure, patients received aspirin (325–500 mg) and a loading dose of clopidogrel (600 mg). Post-PCI antithrombotic therapy included aspirin 200 mg/day continuously and clopidogrel (75 mg/day) for ≥ 1 month if bare-metal stents were implanted or ≥ 6 months if drug-eluting stents were used. At discharge, statins were prescribed to all patients. Other medications were left at the discretion of the attending physician.
2.3
Biochemical measurements
Blood samples were obtained before angiography. Blood was collected into tubes containing lithium–heparin as anticoagulant. Cholesterol and triglycerides were determined in the lithium–heparin plasma samples using standard enzymatic, colorimetric methods. The measuring range was 3.87–800 mg/dl (0.1–20.7 mmol/L) for cholesterol and 8.85–885 mg/dl (0.1–10.0 mmol/L) for triglycerides. LDL-C and HDL-C were measured using homogeneous enzymatic colorimetric assays. The measuring range was 3.86–548 mg/dl (0.10–14.2 mmol/L) for LDL-C and 3–120 mg/dl (0.08–3.10 mmol/L) for HDL-C. In patients with triglyceride concentrations < 400 mg/dl, LDL-C was calculated according to the Friedewald’s formula. All assays were run on a Cobas Integra 800 analyzer (Roche Diagnostics). Plasma concentrations of C-reactive protein were measured using a fully automated, high sensitivity, latex-enhanced immunoturbidometric assay on a Cobas Integra (Roche Diagnostics, Mannheim, Germany). The C-reactive protein assay has a high analytical sensitivity of 0.085 mg/L and a measuring range up to 160 mg/L. Creatinine was measured using a kinetic colorimetric assay based on the compensated Jaffe method. Laboratory personnel involved in the biochemical measurements were unaware of clinical or follow-up data of the patients.
2.4
Outcomes and follow-up
The primary outcome was all-cause 1-year mortality. Secondary outcomes included: cardiac mortality, non-cardiac mortality, non-fatal myocardial infarction, stroke and stent thrombosis. Information on deaths was obtained from hospital records, death certificates or phone contact with the referring physician(s), relatives of the patient, insurance companies and registration of address office. Cardiac death was defined according to the Academic Research Consortium criteria . Non-fatal myocardial infarction was diagnosed in case of development of new abnormal Q waves in ≥ 2 contiguous precordial leads or ≥ 2 adjacent limb leads; or an elevation of creatine kinase-myocardial band (CK-MB) > 2 times (or > 3 times for the 48 h after PCI) the upper limit of normal. The diagnosis of stroke required the development of acute neurologic deficits that were confirmed by computed tomography or magnetic resonance imaging of the head. Definite stent thrombosis was defined using the Academic Research Consortium criteria . The follow-up protocol included a telephone interview at 1 month, a clinic visit at 6 months and a telephone interview at 12 months after the index event. In addition, if patients had cardiac complaints any time during the follow-up a complete clinical, electrocardiographic, and laboratory evaluation was done. Follow-up data were collected by personnel unaware of CRP or plasma lipid levels.
2.5
Statistical analysis
Data are presented as median (25th to 75th percentiles) or counts and proportions (%). One-sample Kolmogorov–Smirnov test was used to assess the normality of data distribution. Categorical data were compared with chi-square test. Continuous data were compared with the Kruskal–Wallis rank-sum test. Survival analysis was performed by applying the Kaplan–Meier method and log-rank test. Multivariable Cox proportional hazards model was used to assess the association between LDL-C and/or CRP with the outcome while adjusting for other potential confounders. All parameters of Table 1 (with the exception of serum creatinine) were entered into the model. CRP was entered after logarithmic transformation and the risk estimates (in this case, hazard ratio) were calculated per standard deviation increase in the CRP logarithmic scale. The discriminatory power of the multivariable models regarding mortality was assessed by calculating the Harrell’s c statistic before and after inclusion of LDL-C, CRP or both and by calculating the integrated discrimination improvement (IDI) . Comparison between c statistics of the models before and after inclusion of LDL-C, CRP, or both alongside other baseline variables was enabled by the Somer’s D package. All analyses were performed using the R statistical package. A 2-tailed P < 0.05 was considered to indicate statistical significance.
