- Email: [email protected]
Factors predicting cardiovascular events in statin-treated diabetic and non-diabetic patients with coronary atherosclerosis
Atherosclerosis 208 (2010) 484–489
Contents lists available at ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
Factors predicting cardiovascular events in statin-treated diabetic and non-diabetic patients with coronary atherosclerosis Heinz Drexel a,b,c,d,∗ , Stefan Aczel a,b , Thomas Marte a,b , Alexander Vonbank a,c , Christoph H. Saely a,b,c a
Vorarlberg Institute for Vascular Investigation and Treatment (VIVIT), Feldkirch, Austria Department of Medicine, Academic Teaching Hospital Feldkirch, Feldkirch, Austria University for Human Sciences, Private University of the Principality of Liechtenstein, Triesen, Liechtenstein d Drexel University College of Medicine, Philadelphia, PA, USA b c
a r t i c l e
i n f o
Article history: Received 8 July 2009 Accepted 12 August 2009 Available online 21 August 2009 Keywords: HDL cholesterol Triglycerides Diabetes Coronary artery disease Prognosis
a b s t r a c t Objective: We aimed at identifying which lipid factors drive vascular risk in statin-treated patients with coronary artery disease (CAD). Methods: We recorded vascular events over 5.6 years in 491 consecutive statin-treated patients with angiographically proven stable CAD, covering 2750 patient-years. Results: In the total population, low high-density lipoprotein (HDL) cholesterol (standardized adjusted HR 0.73 [0.60–0.89]; p = 0.001), low apolipoprotein A1 (0.77 [0.65–0.92]; p = 0.003), a small low-density lipoprotein (LDL) particle diameter (0.76 [0.64–0.91]; p = 0.002), and high triglycerides (1.20 [1.05–1.38]; p = 0.007) predicted vascular events, but not total cholesterol, LDL cholesterol, or apolipoprotein B. Factor analysis in the lipid profiles of our patients revealed an HDL-related factor and an LDL-related factor. Concordant with the results for individual lipid parameters, the HDL-related factor (0.69 [0.58–0.83]; p < 0.001) but not the LDL-related factor (p = 0.455) predicted vascular events. Patients with type 2 diabetes (T2DM; n = 116) were at a higher vascular risk than non-diabetic subjects (38.6% vs. 24.1%; p < 0.001), and like in the total population the HDL-related factor (0.59 [0.44–0.77]; p < 0.001) but not the LDL-related factor (p = 0.591) predicted vascular risk in diabetic patients. Conclusions: The pattern of low HDL cholesterol, low apolipoprotein A1, small LDL particles, and high triglycerides drives vascular risk in statin-treated coronary patients, particularly in those with T2DM. © 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Atherosclerosis is a progressive disease. Eventually, it becomes aggressive and malignant, particularly in patients with type 2 diabetes mellitus (T2DM). As a consequence, atherosclerotic cardiovascular disease is the major cause of death and disability in the general population of industrialized countries, and is responsible for the majority of deaths in patients with T2DM [1]. Prevention of disability and death in these patients can be achieved by halting the progression of atherosclerosis. Antihyperglycemic treatment schedules have failed to reduce cardiovascular disease in patients with T2DM [2,3]. Low-density lipoprotein (LDL) lowering therapy has been much more successful: statins have become a cornerstone in the treatment of atherosclerotic cardiovascular disease, in particular of coronary artery disease (CAD).
∗ Corresponding author at: Department of Medicine and Cardiology, and VIVIT Institute, Academic Teaching Hospital Feldkirch, Carinagasse 47, A-6807 Feldkirch, Austria. Tel.: +43 5522 303 2670; fax: +43 5522 303 7533. E-mail address: [email protected] (H. Drexel). 0021-9150/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2009.08.026
However, only about 30% of cardiovascular events observed with placebo treatment can be prevented by statins [4]. This means that two out of three complications still occur despite statin treatment; even if maximally tolerated doses are used in high risk patients such as those with acute coronary syndromes, only a further 16% of events are prevented. Among the very-high risk patients with both CAD and T2DM, residual risk is particularly unsatisfactory [5]. Statins therefore cannot be the sole solution. The effect of statins on cardiovascular event reduction is closely related to their efficacy in lowering low-density lipoprotein cholesterol [4]. Intervention on other risk factors or pathophysiologic pathways may add to the beneficial effects of statins. Knowledge of the factors driving atherosclerosis in statin-treated patients should be a clue to a further reduction of cardiovascular events. Data from epidemiologic [6] and interventional [7,8] studies point to a role for low high-density lipoprotein (HDL) cholesterol in cardiovascular disease progression. However, no observational data are available to discern which lipid factors predict cardiovascular events in statin-treated CAD patients under contemporary real-life conditions including interventional cardiology and cardiac surgery. Specifically, it is unknown which lipid risk factors or pathophysio-
H. Drexel et al. / Atherosclerosis 208 (2010) 484–489
logic pathways are most important in the very-high risk population of statin-treated patients with T2DM. We therefore undertook an observational 6-year prospective cohort study enrolling diabetic and non-diabetic patients with angiographically proven CAD who received statins to elucidate the factors driving atherosclerosis progression in these patients.
485
2.4. Prospective study
This is a prospective observational investigation on consecutive unselected patients undergoing coronary angiography for the evaluation of CAD. The design of this study has been described in detail earlier [9]. In brief, we enrolled 756 consecutive Caucasian patients referred to coronary angiography for routine evaluation of suspected or established stable CAD. Patients who had suffered myocardial infarctions or acute coronary syndromes within 3 months prior to the baseline angiography were not enrolled. Six patients with type 1 diabetes (C-peptide negative) were excluded from the analyses. For the present analysis, we included the 491 patients of our cohort in whom CAD was proven angiographically and who received statin treatment during a 6-year follow-up. The study complies with the declaration of Helsinki, it was approved by the Ethics Committee of the University of Innsbruck, Austria, and all participants gave written informed consent.
We here report a mean follow-up period of 5.6 ± 1.4 years, covering 2750 patient-years. As pre-specified endpoints, fatal and non-fatal cardiovascular events were recorded, including coronary death (fatal myocardial infarction, sudden cardiac death, mortality from congestive heart failure due to CAD); fatal ischemic stroke; non-fatal myocardial infarction; nonfatal ischemic stroke; and need for coronary artery bypass grafting, percutaneous coronary intervention, or non-coronary revascularization. Time and causes of death were regularly obtained from a national survey (Statistik Austria, Vienna, Austria) or from hospital records. We made every effort to completely assess endpoints in our population; we bi-annually conducted standardized interviews at our institution, and additionally, hospital records of our patients were reviewed. Altogether, we achieved a 100% follow-up rate for mortality data, and completed interviews for 99.6% of our patients. Myocardial infarction was diagnosed according to current guidelines [14]. Stroke was defined as a neurological deficit lasting longer than 48 h with a confirmative computer tomography or magnetic resonance image. Angioplasty and vascular surgery were considered as endpoints unless they were planned as a consequence of the baseline angiography and therefore were not “future” events. All potential endpoints were verified independently by two trained physicians (HD and CHS) blinded to metabolic data including lipid parameters.
2.2. Baseline evaluation
2.5. Statistical analysis
At baseline, coronary angiography was performed with the Judkins technique. CAD was diagnosed in the presence of any lumen narrowing at coronary angiography, and coronary stenoses with lumen narrowing of ≥50% were classified as significant stenoses, as described previously [10]. Hypertension was defined according to the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure [11]; diabetes mellitus was diagnosed according to WHO criteria [12]. Among patients with diabetes, 36.9% were not receiving any anti-diabetic medication, and 42.3%, 32.4%, 27.0%, and 1.8% were receiving – alone or in combination – sulfonylurea, biguanides, insulin, and alpha-glucosidase inhibitors, respectively. Overall, 71.1% of our patients were on aspirin, 12.0% on calcium antagonists, 52.3% on beta adrenoreceptor blocking agents, 37.1% on angiotensin converting enzyme inhibitors, and 4.3% on angiotensin II receptor blocking agents. During follow-up, 6.1%, 0.8%, and 1.0% of our patients besides statin treatment received fibrates, niacin, or ezetimibe. Statin treatment encompassed simvastatin, pravastatin, atorvastatin, fluvastatin, or rosuvastatin.
Differences in baseline characteristics were tested for statistical significance with the Chi-square and the Kruskal–Wallis tests for categorical and continuous variables, respectively. Factor analysis by the method of principal components was applied to extract initial factors from the lipid profiles of our patients. Only factors with an eigenvalue >1 were retained in the analysis. The initial factors were subjected to Varimax rotation to facilitate their interpretation and then were introduced as continuous variables in further analyses. p-values for trend of continuous variables over time were calculated with the Friedman test. The Wilcoxon–Gehan statistic was used to compare differences in the cumulative incidence rates of vascular events. Adjusted hazard ratios (HR) for the incidence of vascular events were derived from Cox proportional hazards models. For these analyses continuous variables were ztransformed and standardized HR were calculated, indicating the change of relative hazards associated with a one unit change of the z-transformed variable, i.e. with one standard deviation of the original variable. Significance was defined as a two-tailed pvalue < 0.05. Results are given as mean ± standard deviation if not denoted otherwise. All statistical analyses were performed with the software package SPSS 11.0 for Windows (SPSS Inc., Chicago, IL, USA).
2. Methods 2.1. Patients and basic study design
2.3. Laboratory analyses Venous blood samples were collected after an overnight fast of 12 h before angiography was performed and laboratory analyses were carried out as described previously [13]. The serum levels of triglycerides, total cholesterol, low-density lipoprotein cholesterol, and high-density lipoprotein cholesterol were determined by using enzymatic hydrolysis and precipitation techniques (Triglycerides GPO-PAP, CHOD/PAP, QuantolipLDL, QuantolipHDL; Roche, Switzerland) on a Hitachi-Analyzer 717 or 911. The LDL peak particle diameter was measured by polyacrylamide gradient gel electrophoresis (Labomed, Germany).
3. Results 3.1. Patient characteristics Baseline data of our cohort were characteristic for a cohort of angiographied coronary patients, with a mean age of 63 ± 10 years, a preponderance of male gender (72.7%), and a high prevalence of hypertension (55.0%), history of smoking (62.9%) and T2DM (23.6%). Baseline characteristics of our patients with normal fasting glucose (NFG), impaired fasting glucose (IFG), and T2DM are summarized in Table 1.
486
H. Drexel et al. / Atherosclerosis 208 (2010) 484–489
Table 1 Baseline characteristics of the study population.
Age (years) Male gender (%) Body mass index (kg/m2 ) Significant coronary stenoses ≥50% (%) Hypertension (%) Smoking (%) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Fasting plasma glucose (mg/dl) HbA1c (%) Triglycerides (mg/dl) Total cholesterol (mg/dl) LDL cholesterol (mg/dl) HDL cholesterol (mg/dl) LDL peak particle diameter (Å) Apolipoprotein A1 (mg/dl) Apolipoprotein B (mg/dl) Impaired renal function (GFR <60 ml/min × 1.73 m2 ) (%) Physical exercise ≥30 min ≥ 2 times/week (%) Prior myocardial infarction (%) Positive family history of CAD (%)
NFG (n = 164)
IFG (n = 211)
T2DM (n = 116)
p-value
63 ± 11 68 26.0 ± 3.5 76 50 58 135 ± 24 78 ± 12 92 ± 5 5.7 ± 0.5 150 ± 81 220 ± 42 135 ± 35 49 ± 12 259 ± 5 146 ± 26 114 ± 24 30 64 35 43
63 ± 10 78 27.3 ± 3.9 79 54 54 134 ± 21 78 ± 11 113 ± 15 5.8 ± 0.5 174 ± 103 221 ± 41 134 ± 35 47 ± 13 258 ± 6 143 ± 25 116 ± 25 21 56 36 39
63 ± 10 69 28.2 ± 4.3 80 65 69 140 ± 21 79 ± 13 161 ± 56 7.6 ± 1.5 213 ± 143 210 ± 49 121 ± 38 43 ± 13 257 ± 6 140 ± 28 112 ± 26 33 52 41 51
0.848 0.695 <0.001 0.419 0.019 0.058 0.077 0.887 <0.001 <0.001 <0.001 0.030 0.001 <0.001 <0.001 0.160 0.160 0.796 0.035 0.283 0.284
NFG, normal fasting glucose; IFG, impaired fasting glucose; T2DM, type 2 diabetes; GFR, glomerular filtration rate; CAD, coronary artery disease. To convert values for fasting plasma glucose to mmol/l multiply by 0.0555, to convert values for triglycerides to mmol/l multiply by 0.0113, and to convert values for total cholesterol, LDL cholesterol, or HDL cholesterol to mmol/l multiply by 0.0259.
3.2. Prospective study Over a mean follow-up period of 5.6 ± 1.4 years, we recorded 161 first vascular events, which thus occurred in 32.8% of our patients, corresponding to an annual event rate of 5.9%. First events included 41 vascular deaths, 22 non-fatal myocardial infarctions, 17 non-fatal strokes, 44 percutaneous coronary interventions, 18 bypass graftings, and 19 revascularizations of non-coronary arteries. In total, we recorded 45 fatal vascular events (including those preceded by a non-fatal event). Table 2 summarizes the results of Cox regression analysis addressing the association between baseline lipid values and future vascular events. After adjustment for age and gender (model 1), HDL cholesterol, apolipoprotein A1, triglycerides, and the LDL particle diameter strongly and significantly predicted vascular events in our cohort of statin-treated CAD patients. These results proved robust after additional adjustment for BMI, smoking and hypertension (model 2). Also after further adjustment for the presence of diabetes (model 3), the same four parameters, low HDL cholesterol, low apolipoprotein A1, a small LDL peak particle diameter, and high triglycerides significantly predicted future vascular events, although with slightly attenuated hazard ratios. In contrast, total cholesterol, LDL cholesterol, and apolipoprotein B were not significantly associated with the incidence of vascular events (Table 2) in any of the models. 3.3. Factor analysis By factor analysis, we extracted two factors from the baseline lipid profiles of our patients. Factor 1 explained 39.2% and factor 2 explained an additional 36.9% of the variance in the lipid variables measured. Correlation coefficients for the correlation between factor 1 and triglycerides, total cholesterol, LDL cholesterol, HDL cholesterol, apolipoprotein A1, apolipoprotein B, and LDL peak particle diameter were 0.284, 0.955, 0.900, 0.222, 0.299, 0.893, and −0.001, respectively. For factor 2, the respective correlation coefficients were −0.698, 0.113, 0.197, 0.887, 0.728, −0.192, and 0.830. Thus, total cholesterol, LDL cholesterol, and apolipoprotein B loaded high on factor 1, which therefore weights LDL (“LDL-related factor”); triglycerides, HDL cholesterol, apolipoprotein A1, and LDL
peak particle diameter loaded high on factor 2, which therefore weights HDL (“HDL-related factor”). Concordant with the results for the individual constituents of these lipid factors, the HDL-related factor (HR = 0.75 [95% CI 0.64–0.89]; p = 0.001) but not the LDL-related factor (HR = 0.97 [0.82–1.14]; p = 0.695) was significantly associated with the incidence of vascular events in univariate analyses. Also, after multivariate adjustment, the HDL-related factor but not the LDLrelated factor predicted vascular events (Fig. 1A). 3.4. Patients with diabetes mellitus From our patients, 164 (33.4%) had NFG, 211 (43.0%) IFG, and 116 (23.6%) T2DM. Vascular risk was significantly higher in patients with T2DM than in non-diabetic subjects: the event rates were 29.3% and 28.9%, and 44.8% in subjects with NFG, in patients with IFG, and in patients with T2DM, respectively (ptrend = 0.007). Like in the total study population, low HDL cholesterol, low apolipoprotein A, high triglycerides, and a small LDL particle diameter significantly predicted vascular events in patients with T2DM (Table 2). As can be seen from Fig. 1B, the HDL-related factor was a particularly strong predictor of vascular events in patients with T2DM. In the patient subgroups with IFG and NFG its association with vascular events did not reach statistical significance when these subgroups were considered separately. However, when in order to increase statistical power, IFG and NFG subjects were pooled into a single non-diabetic group, the HDL-related factor proved significantly predictive of vascular events also among nondiabetic patients (standardized HR adjusted for age, gender, BMI smoking, and hypertension = 0.73 [0.57–0.93]; p = 0.011). The LDLrelated factor neither among patients with T2DM (see Fig. 1B) nor among non-diabetic patients (standardized adjusted HR = 1.09 [0.89–1.33]; p = 0.400) was associated with vascular events. 4. Discussion With our 6-year observational study we were able to identify the pattern of high triglycerides, small dense LDL, low HDL cholesterol and low apolipoprotein A1 as the paramount predictor of disease progression in statin-treated CAD patients. In particular, this pat-
H. Drexel et al. / Atherosclerosis 208 (2010) 484–489
487
Table 2 Results of Cox regression analyses. Total study population
NGT
IFG
T2DM
HDL cholesterol
Model 1 Model 2 Model 3
0.73 [0.60–0.88]; p = 0.001 0.70 [0.58–0.85]; p <0.001 0.73 [0.60–0.89]; p = 0.001
0.76 [0.53–1.08]; p = 0.127 0.69 [0.48–0.99]; p = 0.042 n.a.
0.93 [0.69–1.24]; p = 0.605 0.89 [0.66–1.21]; p = 0.447 n.a.
0.60 [0.43–0.85]; p = 0.004 0.58 [0.41–0.82]; p = 0.002 n.a.
Triglycerides
Model 1 Model 2 Model 3
1.23 [1.08–1.41]; p = 0.002 1.25 [1.10–1.43]; p = 0.001 1.20 [1.05–1.38]; p = 0.007
0.82 [0.51–1.32]; p = 0.412 0.85 [0.51–1.40]; p = 0.523 n.a.
1.18 [0.86–1.61]; p = 0.303 1.26 [0.92–1.74]; p = 0.155 n.a.
1.26 [1.08–1.46]; p = 0.004 1.30 [1.11–1.53]; p = 0.001 n.a.
Apo A1
Model 1 Model 2 Model 3
0.77 [0.65–0.91]; p = 0.002 0.76 [0.63–0.90]; p = 0.001 0.77 [0.65–0.92]; p = 0.003
0.79 [0.58–1.08]; p = 0.138 0.76 [0.56–1.03]; p = 0.074 n.a.
0.81 [0.61–1.07]; p = 0.128 0.80 [0.60–1.06]; p = 0.115 n.a.
0.73 [0.54–0.97]; p = 0.033 0.70 [0.51–0.95]; p = 0.022 n.a.
LDL-PPD
Model 1 Model 2 Model 3
0.76 [0.64–0.90]; p = 0.001 0.74 [0.62–0.88]; p = 0.001 0.76 [0.64–0.91]; p = 0.002
0.85 [0.60–1.20]; p = 0.344 0.82 [0.58–1.17]; p = 0.278 n.a.
0.84 [0.63–1.11]; p = 0.216 0.78 [0.58–1.05]; p = 0.108 n.a.
0.70 [0.53–0.93]; p = 0.012 0.67 [0.50–0.91]; p = 0.010 n.a.
Total cholesterol
Model 1 Model 2 Model 3
0.97 [0.82–1.13]; p = 0.661 0.97 [0.83–1.14]; p = 0.728 1.01 [0.86–1.18]; p = 0.955
1.05 [0.77–1.42]; p = 0.760 1.02 [0.75–1.39]; p = 0.878 n.a.
0.95 [0.72–1.27]; p = 0.732 0.97 [0.73–1.30]; p = 0.859 n.a.
1.02 [0.79–1.29]; p = 0.941 1.03 [0.79–1.35]; p = 0.822 n.a.
LDL cholesterol
Model 1 Model 2 Model 3
0.96 [0.81–1.12]; p = 0.572 0.97 [0.82–1.13]; p = 0.664 1.00 [0.86–1.18]; p = 0.961
1.21 [0.90–1.63]; p = 0.200 1.20 [0.89–1.62]; p = 0.232 n.a.
1.02 [0.77–1.34]; p = 0.914 1.03 [0.79–1.36]; p = 0.811 n.a.
0.83 [0.63–1.10]; p = 0.190 0.85 [0.64–1.12]; p = 0.235 n.a.
Apo B
Model 1 Model 2 Model 3
1.11 [0.95–1.30]; p = 0.197 1.14 [0.97–1.33]; p = 0.112 1.15 [0.99–1.34]; p = 0.077
1.16 [0.87–1.56]; p = 0.310 1.16 [0.87–1.56]; p = 0.317 n.a.
1.14 [0.88–1.47]; p = 0.319 1.22 [0.94–1.58]; p = 0.138 n.a.
1.10 [0.85–1.41]; p = 0.469 1.13 [0.87–1.48]; p = 0.366 n.a.
Apo B/A1 ratio
Model 1 Model 2 Model 3
1.24 [1.07–1.44]; p = 0.005 1.27 [1.10–1.48]; p = 0.002 1.26 [1.09–1.46]; p = 0.002
1.27 [0.92–1.74]; p = 0.143 1.27 [0.93–1.73]; p = 0.129 n.a.
2.20 [0.95–1.52]; p = 0.124 1.29 [1.01–1.65]; p = 0.039 n.a.
1.28 [1.00–1.63]; p = 0.050 1.32 [1.03–1.70]; p = 0.028 n.a.
Model 1: adjusted for age and gender; model 2: adjusted for age, gender, body mass index, smoking, and hypertension; model 3: adjusted for the covariates included in model 2, and additionally for type 2 diabetes. Standardized adjusted hazard ratios together with the 95% confidence intervals (CI) are given. Apo A1, apolipoprotein A1; Apo B, apolipoprotein B; LDL, low-density lipoprotein; LDL-PPD, LDL peak particle diameter; HDL, high-density lipoprotein.
Fig. 1. Lipid factors derived from factor analysis as predictors of vascular events. (A) Total study population – model 1: adjusted for age and gender; model 2: adjusted for age, gender, body mass index, smoking, and hypertension; model 3: adjusted for the covariates included in model 2, and additionally for type 2 diabetes. Standardized adjusted hazard ratios [95% confidence intervals] are given. (B) Subgroup analyses with respect to the glycemic state – standardized hazard ratios [95% confidence intervals] are given adjusted for age, gender, body mass index, smoking, and hypertension. LDL, low-density lipoprotein; HDL, high-density lipoprotein; HR, hazard ratio; NFG, normal fasting glucose; IFG, impaired fasting glucose; T2DM, type 2 diabetes.
488
H. Drexel et al. / Atherosclerosis 208 (2010) 484–489
tern strongly predicted future cardiovascular events in patients with manifest T2DM. Pathophyisologically, high triglycerides, low HDL cholesterol, low apolipoprotein A1, and small dense LDL are four different facets of a common metabolic pathway which is driven by insulin resistance [15]. Together, these parameters constitute a lipid pattern which drives accelerated progression of atherosclerotic vascular disease. Because predictive power is lost when only single components from a pattern are considered, the use of an integrative factor should increase discriminative strength. Therefore, it appears adequate to address the comprehensive risk conferred by a pattern of lipid abnormalities by factor analysis, a mathematical technique that integrates common aspects of several individual variables into a single new variable. Indeed, with this novel approach, in our study the HDL-related lipid factor robustly predicted vascular events. The LDL-related factor – which in a similar way integrated common properties of total cholesterol, LDL cholesterol, and apolipoprotein B – in contrast was not associated with vascular events. Importantly, the low HDL cholesterol, high triglycerides, small LDL particle pattern of dyslipidemia was most pronounced and most strongly predicted vascular risk in patients with T2DM. Statins predominantly decrease LDL cholesterol; the risk reduction achieved with statins may therefore be considered to be primarily due to lowering of LDL cholesterol. Because statins only marginally (5–8%) raise HDL cholesterol [16], our data support the notion that the paramount risk indicated by low HDL cholesterol is not adequately addressed by statin therapy. Here the question arises how this risk could be intervened. Life style changes have the potential to improve dyslipidemia through interaction at the underlying disorders of central obesity and insulin resistance, which themselves strongly predict future cardiovascular events as we have shown in a similar population [9,17]. In the light of our data, a focus on more aggressive life style interventions that improve triglycerides, HDL cholesterol, and small dense LDL, e.g. smoking cessation, dietary and exercise counselling appears mandatory. Perhaps the success with drugs such as statins has detracted our attention too much from life style modification of lipids. Among pharmacological interventions, nicotinic acid is a powerful compound to improve triglycerides, HDL cholesterol, small dense LDL, and in addition the independent risk factor lipoprotein(a). There is some evidence from trials that an intervention with nicotinic acid is beneficial in combination with statins [18,19]. The AIM-HIGH [20] and HPS-2 THRIVE [21] trials currently test the potential of a statin-nicotinic acid combination to reduce cardiovascular risk over and above statin monotherapy. Because of the disappointing results of the FIELD study [22], fibrates at present appear less promising, but the results of the fenofibrate arm of the ACCORD trial [23] are still being awaited. Torcetrapib, an inhibitor of CETP that strongly increases HDL cholesterol, did not meet the high expectations [24]. The strengths of our study include four aspects: first, the observational design ascertained that we investigated patients under real-world conditions without a bias from patient selection criteria inherent in interventional studies. Second, CAD was verified by angiography in every single patient. Third, we achieved an almost complete follow-up (99.6%), which is exceptionally high for an observational study. Fourth, we have investigated the power of a broad array of lipids, lipoproteins, and apolipoproteins as predictors of cardiovascular events and integrated them into factor analysis. Taken together, our population is representative, well characterised, and meticulously monitored. As a possible limitation of our investigation, despite coverage of about 2750 patient-years, the sample size warrants consideration. From our data we cannot exclude that, with the greater statistical power of a larger study or a longer follow-up period, an associ-
ation between weaker risk factors and vascular events could be demonstrable. In conclusion, our results suggest that after LDL lowering with statins, HDL raising/triglyceride lowering therapy should be considered as the next important step in lipid intervention for CAD patients, particularly in those with T2DM. Conflict of interest None declared. Acknowledgements The VIVIT Institute thanks Dr. Egmond Frommelt and the Innovationsstiftung of the Liechtenstein Global Trust (LGT) Bank (Bendern, Liechtenstein), Dr. Karl Josef Hier and the Peter Goop Stiftung (Vaduz, Liechtenstein), the Fachhochschule Dornbirn (Dornbirn, Austria), and the Institute for Clinical Chemistry at the Academic Teaching Hospital Feldkirch (Feldkirch, Austria) for providing us with generous research grants. We are grateful to Franz Rauch and the Vorarlberger Industriellenvereinigung (Bregenz, Austria), to Dr. Peter Woess and the Vorarlberger Aerztekammer (Dornbirn, Austria), to Mag. Gabriela Duer and the Vorarlberger Landesregierung, to Dr. Elmar Bechter, Landessanitätsdirektor (Bregenz, Austria) and to Luis Patsch, former Director, Vorarlberger Landeskrankenhaus-Betriebsgesellschaft, for continuously supporting our Research Institute. References [1] McEwen LN, Kim C, Karter AJ, et al. Risk factors for mortality among patients with diabetes: the Translating Research Into Action for Diabetes (TRIAD) study. Diabetes Care 2007;30:1736–41. [2] Cefalu WT. Glycemic targets and cardiovascular disease. N Engl J Med 2008;358:2633–5. [3] Dluhy RG, McMahon GT. Intensive glycemic control in the ACCORD and ADVANCE trials. N Engl J Med 2008;358:2630–3. [4] Baigent C, Keech A, Kearney PM, et al. Efficacy and safety of cholesterollowering treatment: prospective meta-analysis of data from 90,056 participants in 14 randomised trials of statins. Lancet 2005;366:1267–78. [5] Shepherd J, Barter P, Carmena R, et al. Effect of lowering LDL cholesterol substantially below currently recommended levels in patients with coronary heart disease and diabetes: the Treating to New Targets (TNT) study. Diabetes Care 2006;29:1220–6. [6] Castelli WP, Garrison RJ, Wilson PW, et al. Incidence of coronary heart disease and lipoprotein cholesterol levels. The Framingham Study. JAMA 1986;256:2835–8. [7] Barter P, Gotto AM, LaRosa JC, et al. HDL cholesterol, very low levels of LDL cholesterol, and cardiovascular events. N Engl J Med 2007;357:1301–10. [8] Amarenco P, Goldstein LB, Callahan III A, et al. Baseline blood pressure, low- and high-density lipoproteins, and triglycerides and the risk of vascular events in the Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) trial. Atherosclerosis 2009;204:515–20. [9] Saely CH, Aczel S, Marte T, et al. The metabolic syndrome, insulin resistance, and cardiovascular risk in diabetic and nondiabetic patients. J Clin Endocrinol Metab 2005;90:5698–703. [10] Drexel H, Amann FW, Beran J, et al. Plasma triglycerides and three lipoprotein cholesterol fractions are independent predictors of the extent of coronary atherosclerosis. Circulation 1994;90:2230–5. [11] Chobanian AV, Bakris GL, Black HR, et al. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA 2003;289:2560–72. [12] Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1. Diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabet Med 1998;15:539–53. [13] Drexel H, Aczel S, Marte T, et al. Is atherosclerosis in diabetes and impaired fasting glucose driven by elevated LDL cholesterol or by decreased HDL cholesterol? Diabetes Care 2005;28:101–7. [14] Alpert JS, Thygesen K, Antman E, Bassand JP. Myocardial infarction redefined—a consensus document of The Joint European Society of Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction. J Am Coll Cardiol 2000;36:959–69. [15] Taskinen MR. Diabetic dyslipidaemia: from basic research to clinical practice. Diabetologia 2003;46:733–49. [16] Schaefer EJ, Asztalos BF. The effects of statins on high-density lipoproteins. Curr Atheroscler Rep 2006;8:41–9.
H. Drexel et al. / Atherosclerosis 208 (2010) 484–489 [17] Hoefle G, Saely CH, Aczel S, et al. Impact of total and central obesity on vascular mortality in patients undergoing coronary angiography. Int J Obes (Lond) 2005;29:785–91. [18] Brown BG, Zhao XQ, Chait A, et al. Simvastatin and niacin, antioxidant vitamins, or the combination for the prevention of coronary disease. N Engl J Med 2001;345:1583–92. [19] Taylor AJ, Sullenberger LE, Lee HJ, Lee JK, Grace KA. Arterial biology for the investigation of the treatment effects of reducing cholesterol (ARBITER) 2: a double-blind, placebo-controlled study of extended-release niacin on atherosclerosis progression in secondary prevention patients treated with statins. Circulation 2004;110:3512–7. [20] AIM HIGH: Niacin Plus Statin to Prevent Vascular Events. www.clinicaltrials. gov, 2009 (accessed 07.07.09).
489
[21] Treatment of HDL to Reduce the Incidence of Vascular Events HPS2-THRIVE. www.clinicaltrials.gov, 2009 (accessed 07.07.09). [22] Keech A, Simes RJ, Barter P, et al. Effects of long-term fenofibrate therapy on cardiovascular events in 9795 people with type 2 diabetes mellitus (the FIELD study): randomised controlled trial. Lancet 2005;366:1849– 61. [23] Ginsberg HN, Bonds DE, Lovato LC, et al. Evolution of the lipid trial protocol of the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial. Am J Cardiol 2007;99:56i–67i. [24] Tall AR, Yvan-Charvet L, Wang N. The failure of torcetrapib: was it the molecule or the mechanism? Arterioscler Thromb Vasc Biol 2007;27:257–60.
Contents lists available at ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
Factors predicting cardiovascular events in statin-treated diabetic and non-diabetic patients with coronary atherosclerosis Heinz Drexel a,b,c,d,∗ , Stefan Aczel a,b , Thomas Marte a,b , Alexander Vonbank a,c , Christoph H. Saely a,b,c a
Vorarlberg Institute for Vascular Investigation and Treatment (VIVIT), Feldkirch, Austria Department of Medicine, Academic Teaching Hospital Feldkirch, Feldkirch, Austria University for Human Sciences, Private University of the Principality of Liechtenstein, Triesen, Liechtenstein d Drexel University College of Medicine, Philadelphia, PA, USA b c
a r t i c l e
i n f o
Article history: Received 8 July 2009 Accepted 12 August 2009 Available online 21 August 2009 Keywords: HDL cholesterol Triglycerides Diabetes Coronary artery disease Prognosis
a b s t r a c t Objective: We aimed at identifying which lipid factors drive vascular risk in statin-treated patients with coronary artery disease (CAD). Methods: We recorded vascular events over 5.6 years in 491 consecutive statin-treated patients with angiographically proven stable CAD, covering 2750 patient-years. Results: In the total population, low high-density lipoprotein (HDL) cholesterol (standardized adjusted HR 0.73 [0.60–0.89]; p = 0.001), low apolipoprotein A1 (0.77 [0.65–0.92]; p = 0.003), a small low-density lipoprotein (LDL) particle diameter (0.76 [0.64–0.91]; p = 0.002), and high triglycerides (1.20 [1.05–1.38]; p = 0.007) predicted vascular events, but not total cholesterol, LDL cholesterol, or apolipoprotein B. Factor analysis in the lipid profiles of our patients revealed an HDL-related factor and an LDL-related factor. Concordant with the results for individual lipid parameters, the HDL-related factor (0.69 [0.58–0.83]; p < 0.001) but not the LDL-related factor (p = 0.455) predicted vascular events. Patients with type 2 diabetes (T2DM; n = 116) were at a higher vascular risk than non-diabetic subjects (38.6% vs. 24.1%; p < 0.001), and like in the total population the HDL-related factor (0.59 [0.44–0.77]; p < 0.001) but not the LDL-related factor (p = 0.591) predicted vascular risk in diabetic patients. Conclusions: The pattern of low HDL cholesterol, low apolipoprotein A1, small LDL particles, and high triglycerides drives vascular risk in statin-treated coronary patients, particularly in those with T2DM. © 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Atherosclerosis is a progressive disease. Eventually, it becomes aggressive and malignant, particularly in patients with type 2 diabetes mellitus (T2DM). As a consequence, atherosclerotic cardiovascular disease is the major cause of death and disability in the general population of industrialized countries, and is responsible for the majority of deaths in patients with T2DM [1]. Prevention of disability and death in these patients can be achieved by halting the progression of atherosclerosis. Antihyperglycemic treatment schedules have failed to reduce cardiovascular disease in patients with T2DM [2,3]. Low-density lipoprotein (LDL) lowering therapy has been much more successful: statins have become a cornerstone in the treatment of atherosclerotic cardiovascular disease, in particular of coronary artery disease (CAD).
∗ Corresponding author at: Department of Medicine and Cardiology, and VIVIT Institute, Academic Teaching Hospital Feldkirch, Carinagasse 47, A-6807 Feldkirch, Austria. Tel.: +43 5522 303 2670; fax: +43 5522 303 7533. E-mail address: [email protected] (H. Drexel). 0021-9150/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2009.08.026
However, only about 30% of cardiovascular events observed with placebo treatment can be prevented by statins [4]. This means that two out of three complications still occur despite statin treatment; even if maximally tolerated doses are used in high risk patients such as those with acute coronary syndromes, only a further 16% of events are prevented. Among the very-high risk patients with both CAD and T2DM, residual risk is particularly unsatisfactory [5]. Statins therefore cannot be the sole solution. The effect of statins on cardiovascular event reduction is closely related to their efficacy in lowering low-density lipoprotein cholesterol [4]. Intervention on other risk factors or pathophysiologic pathways may add to the beneficial effects of statins. Knowledge of the factors driving atherosclerosis in statin-treated patients should be a clue to a further reduction of cardiovascular events. Data from epidemiologic [6] and interventional [7,8] studies point to a role for low high-density lipoprotein (HDL) cholesterol in cardiovascular disease progression. However, no observational data are available to discern which lipid factors predict cardiovascular events in statin-treated CAD patients under contemporary real-life conditions including interventional cardiology and cardiac surgery. Specifically, it is unknown which lipid risk factors or pathophysio-
H. Drexel et al. / Atherosclerosis 208 (2010) 484–489
logic pathways are most important in the very-high risk population of statin-treated patients with T2DM. We therefore undertook an observational 6-year prospective cohort study enrolling diabetic and non-diabetic patients with angiographically proven CAD who received statins to elucidate the factors driving atherosclerosis progression in these patients.
485
2.4. Prospective study
This is a prospective observational investigation on consecutive unselected patients undergoing coronary angiography for the evaluation of CAD. The design of this study has been described in detail earlier [9]. In brief, we enrolled 756 consecutive Caucasian patients referred to coronary angiography for routine evaluation of suspected or established stable CAD. Patients who had suffered myocardial infarctions or acute coronary syndromes within 3 months prior to the baseline angiography were not enrolled. Six patients with type 1 diabetes (C-peptide negative) were excluded from the analyses. For the present analysis, we included the 491 patients of our cohort in whom CAD was proven angiographically and who received statin treatment during a 6-year follow-up. The study complies with the declaration of Helsinki, it was approved by the Ethics Committee of the University of Innsbruck, Austria, and all participants gave written informed consent.
We here report a mean follow-up period of 5.6 ± 1.4 years, covering 2750 patient-years. As pre-specified endpoints, fatal and non-fatal cardiovascular events were recorded, including coronary death (fatal myocardial infarction, sudden cardiac death, mortality from congestive heart failure due to CAD); fatal ischemic stroke; non-fatal myocardial infarction; nonfatal ischemic stroke; and need for coronary artery bypass grafting, percutaneous coronary intervention, or non-coronary revascularization. Time and causes of death were regularly obtained from a national survey (Statistik Austria, Vienna, Austria) or from hospital records. We made every effort to completely assess endpoints in our population; we bi-annually conducted standardized interviews at our institution, and additionally, hospital records of our patients were reviewed. Altogether, we achieved a 100% follow-up rate for mortality data, and completed interviews for 99.6% of our patients. Myocardial infarction was diagnosed according to current guidelines [14]. Stroke was defined as a neurological deficit lasting longer than 48 h with a confirmative computer tomography or magnetic resonance image. Angioplasty and vascular surgery were considered as endpoints unless they were planned as a consequence of the baseline angiography and therefore were not “future” events. All potential endpoints were verified independently by two trained physicians (HD and CHS) blinded to metabolic data including lipid parameters.
2.2. Baseline evaluation
2.5. Statistical analysis
At baseline, coronary angiography was performed with the Judkins technique. CAD was diagnosed in the presence of any lumen narrowing at coronary angiography, and coronary stenoses with lumen narrowing of ≥50% were classified as significant stenoses, as described previously [10]. Hypertension was defined according to the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure [11]; diabetes mellitus was diagnosed according to WHO criteria [12]. Among patients with diabetes, 36.9% were not receiving any anti-diabetic medication, and 42.3%, 32.4%, 27.0%, and 1.8% were receiving – alone or in combination – sulfonylurea, biguanides, insulin, and alpha-glucosidase inhibitors, respectively. Overall, 71.1% of our patients were on aspirin, 12.0% on calcium antagonists, 52.3% on beta adrenoreceptor blocking agents, 37.1% on angiotensin converting enzyme inhibitors, and 4.3% on angiotensin II receptor blocking agents. During follow-up, 6.1%, 0.8%, and 1.0% of our patients besides statin treatment received fibrates, niacin, or ezetimibe. Statin treatment encompassed simvastatin, pravastatin, atorvastatin, fluvastatin, or rosuvastatin.
Differences in baseline characteristics were tested for statistical significance with the Chi-square and the Kruskal–Wallis tests for categorical and continuous variables, respectively. Factor analysis by the method of principal components was applied to extract initial factors from the lipid profiles of our patients. Only factors with an eigenvalue >1 were retained in the analysis. The initial factors were subjected to Varimax rotation to facilitate their interpretation and then were introduced as continuous variables in further analyses. p-values for trend of continuous variables over time were calculated with the Friedman test. The Wilcoxon–Gehan statistic was used to compare differences in the cumulative incidence rates of vascular events. Adjusted hazard ratios (HR) for the incidence of vascular events were derived from Cox proportional hazards models. For these analyses continuous variables were ztransformed and standardized HR were calculated, indicating the change of relative hazards associated with a one unit change of the z-transformed variable, i.e. with one standard deviation of the original variable. Significance was defined as a two-tailed pvalue < 0.05. Results are given as mean ± standard deviation if not denoted otherwise. All statistical analyses were performed with the software package SPSS 11.0 for Windows (SPSS Inc., Chicago, IL, USA).
2. Methods 2.1. Patients and basic study design
2.3. Laboratory analyses Venous blood samples were collected after an overnight fast of 12 h before angiography was performed and laboratory analyses were carried out as described previously [13]. The serum levels of triglycerides, total cholesterol, low-density lipoprotein cholesterol, and high-density lipoprotein cholesterol were determined by using enzymatic hydrolysis and precipitation techniques (Triglycerides GPO-PAP, CHOD/PAP, QuantolipLDL, QuantolipHDL; Roche, Switzerland) on a Hitachi-Analyzer 717 or 911. The LDL peak particle diameter was measured by polyacrylamide gradient gel electrophoresis (Labomed, Germany).
3. Results 3.1. Patient characteristics Baseline data of our cohort were characteristic for a cohort of angiographied coronary patients, with a mean age of 63 ± 10 years, a preponderance of male gender (72.7%), and a high prevalence of hypertension (55.0%), history of smoking (62.9%) and T2DM (23.6%). Baseline characteristics of our patients with normal fasting glucose (NFG), impaired fasting glucose (IFG), and T2DM are summarized in Table 1.
486
H. Drexel et al. / Atherosclerosis 208 (2010) 484–489
Table 1 Baseline characteristics of the study population.
Age (years) Male gender (%) Body mass index (kg/m2 ) Significant coronary stenoses ≥50% (%) Hypertension (%) Smoking (%) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Fasting plasma glucose (mg/dl) HbA1c (%) Triglycerides (mg/dl) Total cholesterol (mg/dl) LDL cholesterol (mg/dl) HDL cholesterol (mg/dl) LDL peak particle diameter (Å) Apolipoprotein A1 (mg/dl) Apolipoprotein B (mg/dl) Impaired renal function (GFR <60 ml/min × 1.73 m2 ) (%) Physical exercise ≥30 min ≥ 2 times/week (%) Prior myocardial infarction (%) Positive family history of CAD (%)
NFG (n = 164)
IFG (n = 211)
T2DM (n = 116)
p-value
63 ± 11 68 26.0 ± 3.5 76 50 58 135 ± 24 78 ± 12 92 ± 5 5.7 ± 0.5 150 ± 81 220 ± 42 135 ± 35 49 ± 12 259 ± 5 146 ± 26 114 ± 24 30 64 35 43
63 ± 10 78 27.3 ± 3.9 79 54 54 134 ± 21 78 ± 11 113 ± 15 5.8 ± 0.5 174 ± 103 221 ± 41 134 ± 35 47 ± 13 258 ± 6 143 ± 25 116 ± 25 21 56 36 39
63 ± 10 69 28.2 ± 4.3 80 65 69 140 ± 21 79 ± 13 161 ± 56 7.6 ± 1.5 213 ± 143 210 ± 49 121 ± 38 43 ± 13 257 ± 6 140 ± 28 112 ± 26 33 52 41 51
0.848 0.695 <0.001 0.419 0.019 0.058 0.077 0.887 <0.001 <0.001 <0.001 0.030 0.001 <0.001 <0.001 0.160 0.160 0.796 0.035 0.283 0.284
NFG, normal fasting glucose; IFG, impaired fasting glucose; T2DM, type 2 diabetes; GFR, glomerular filtration rate; CAD, coronary artery disease. To convert values for fasting plasma glucose to mmol/l multiply by 0.0555, to convert values for triglycerides to mmol/l multiply by 0.0113, and to convert values for total cholesterol, LDL cholesterol, or HDL cholesterol to mmol/l multiply by 0.0259.
3.2. Prospective study Over a mean follow-up period of 5.6 ± 1.4 years, we recorded 161 first vascular events, which thus occurred in 32.8% of our patients, corresponding to an annual event rate of 5.9%. First events included 41 vascular deaths, 22 non-fatal myocardial infarctions, 17 non-fatal strokes, 44 percutaneous coronary interventions, 18 bypass graftings, and 19 revascularizations of non-coronary arteries. In total, we recorded 45 fatal vascular events (including those preceded by a non-fatal event). Table 2 summarizes the results of Cox regression analysis addressing the association between baseline lipid values and future vascular events. After adjustment for age and gender (model 1), HDL cholesterol, apolipoprotein A1, triglycerides, and the LDL particle diameter strongly and significantly predicted vascular events in our cohort of statin-treated CAD patients. These results proved robust after additional adjustment for BMI, smoking and hypertension (model 2). Also after further adjustment for the presence of diabetes (model 3), the same four parameters, low HDL cholesterol, low apolipoprotein A1, a small LDL peak particle diameter, and high triglycerides significantly predicted future vascular events, although with slightly attenuated hazard ratios. In contrast, total cholesterol, LDL cholesterol, and apolipoprotein B were not significantly associated with the incidence of vascular events (Table 2) in any of the models. 3.3. Factor analysis By factor analysis, we extracted two factors from the baseline lipid profiles of our patients. Factor 1 explained 39.2% and factor 2 explained an additional 36.9% of the variance in the lipid variables measured. Correlation coefficients for the correlation between factor 1 and triglycerides, total cholesterol, LDL cholesterol, HDL cholesterol, apolipoprotein A1, apolipoprotein B, and LDL peak particle diameter were 0.284, 0.955, 0.900, 0.222, 0.299, 0.893, and −0.001, respectively. For factor 2, the respective correlation coefficients were −0.698, 0.113, 0.197, 0.887, 0.728, −0.192, and 0.830. Thus, total cholesterol, LDL cholesterol, and apolipoprotein B loaded high on factor 1, which therefore weights LDL (“LDL-related factor”); triglycerides, HDL cholesterol, apolipoprotein A1, and LDL
peak particle diameter loaded high on factor 2, which therefore weights HDL (“HDL-related factor”). Concordant with the results for the individual constituents of these lipid factors, the HDL-related factor (HR = 0.75 [95% CI 0.64–0.89]; p = 0.001) but not the LDL-related factor (HR = 0.97 [0.82–1.14]; p = 0.695) was significantly associated with the incidence of vascular events in univariate analyses. Also, after multivariate adjustment, the HDL-related factor but not the LDLrelated factor predicted vascular events (Fig. 1A). 3.4. Patients with diabetes mellitus From our patients, 164 (33.4%) had NFG, 211 (43.0%) IFG, and 116 (23.6%) T2DM. Vascular risk was significantly higher in patients with T2DM than in non-diabetic subjects: the event rates were 29.3% and 28.9%, and 44.8% in subjects with NFG, in patients with IFG, and in patients with T2DM, respectively (ptrend = 0.007). Like in the total study population, low HDL cholesterol, low apolipoprotein A, high triglycerides, and a small LDL particle diameter significantly predicted vascular events in patients with T2DM (Table 2). As can be seen from Fig. 1B, the HDL-related factor was a particularly strong predictor of vascular events in patients with T2DM. In the patient subgroups with IFG and NFG its association with vascular events did not reach statistical significance when these subgroups were considered separately. However, when in order to increase statistical power, IFG and NFG subjects were pooled into a single non-diabetic group, the HDL-related factor proved significantly predictive of vascular events also among nondiabetic patients (standardized HR adjusted for age, gender, BMI smoking, and hypertension = 0.73 [0.57–0.93]; p = 0.011). The LDLrelated factor neither among patients with T2DM (see Fig. 1B) nor among non-diabetic patients (standardized adjusted HR = 1.09 [0.89–1.33]; p = 0.400) was associated with vascular events. 4. Discussion With our 6-year observational study we were able to identify the pattern of high triglycerides, small dense LDL, low HDL cholesterol and low apolipoprotein A1 as the paramount predictor of disease progression in statin-treated CAD patients. In particular, this pat-
H. Drexel et al. / Atherosclerosis 208 (2010) 484–489
487
Table 2 Results of Cox regression analyses. Total study population
NGT
IFG
T2DM
HDL cholesterol
Model 1 Model 2 Model 3
0.73 [0.60–0.88]; p = 0.001 0.70 [0.58–0.85]; p <0.001 0.73 [0.60–0.89]; p = 0.001
0.76 [0.53–1.08]; p = 0.127 0.69 [0.48–0.99]; p = 0.042 n.a.
0.93 [0.69–1.24]; p = 0.605 0.89 [0.66–1.21]; p = 0.447 n.a.
0.60 [0.43–0.85]; p = 0.004 0.58 [0.41–0.82]; p = 0.002 n.a.
Triglycerides
Model 1 Model 2 Model 3
1.23 [1.08–1.41]; p = 0.002 1.25 [1.10–1.43]; p = 0.001 1.20 [1.05–1.38]; p = 0.007
0.82 [0.51–1.32]; p = 0.412 0.85 [0.51–1.40]; p = 0.523 n.a.
1.18 [0.86–1.61]; p = 0.303 1.26 [0.92–1.74]; p = 0.155 n.a.
1.26 [1.08–1.46]; p = 0.004 1.30 [1.11–1.53]; p = 0.001 n.a.
Apo A1
Model 1 Model 2 Model 3
0.77 [0.65–0.91]; p = 0.002 0.76 [0.63–0.90]; p = 0.001 0.77 [0.65–0.92]; p = 0.003
0.79 [0.58–1.08]; p = 0.138 0.76 [0.56–1.03]; p = 0.074 n.a.
0.81 [0.61–1.07]; p = 0.128 0.80 [0.60–1.06]; p = 0.115 n.a.
0.73 [0.54–0.97]; p = 0.033 0.70 [0.51–0.95]; p = 0.022 n.a.
LDL-PPD
Model 1 Model 2 Model 3
0.76 [0.64–0.90]; p = 0.001 0.74 [0.62–0.88]; p = 0.001 0.76 [0.64–0.91]; p = 0.002
0.85 [0.60–1.20]; p = 0.344 0.82 [0.58–1.17]; p = 0.278 n.a.
0.84 [0.63–1.11]; p = 0.216 0.78 [0.58–1.05]; p = 0.108 n.a.
0.70 [0.53–0.93]; p = 0.012 0.67 [0.50–0.91]; p = 0.010 n.a.
Total cholesterol
Model 1 Model 2 Model 3
0.97 [0.82–1.13]; p = 0.661 0.97 [0.83–1.14]; p = 0.728 1.01 [0.86–1.18]; p = 0.955
1.05 [0.77–1.42]; p = 0.760 1.02 [0.75–1.39]; p = 0.878 n.a.
0.95 [0.72–1.27]; p = 0.732 0.97 [0.73–1.30]; p = 0.859 n.a.
1.02 [0.79–1.29]; p = 0.941 1.03 [0.79–1.35]; p = 0.822 n.a.
LDL cholesterol
Model 1 Model 2 Model 3
0.96 [0.81–1.12]; p = 0.572 0.97 [0.82–1.13]; p = 0.664 1.00 [0.86–1.18]; p = 0.961
1.21 [0.90–1.63]; p = 0.200 1.20 [0.89–1.62]; p = 0.232 n.a.
1.02 [0.77–1.34]; p = 0.914 1.03 [0.79–1.36]; p = 0.811 n.a.
0.83 [0.63–1.10]; p = 0.190 0.85 [0.64–1.12]; p = 0.235 n.a.
Apo B
Model 1 Model 2 Model 3
1.11 [0.95–1.30]; p = 0.197 1.14 [0.97–1.33]; p = 0.112 1.15 [0.99–1.34]; p = 0.077
1.16 [0.87–1.56]; p = 0.310 1.16 [0.87–1.56]; p = 0.317 n.a.
1.14 [0.88–1.47]; p = 0.319 1.22 [0.94–1.58]; p = 0.138 n.a.
1.10 [0.85–1.41]; p = 0.469 1.13 [0.87–1.48]; p = 0.366 n.a.
Apo B/A1 ratio
Model 1 Model 2 Model 3
1.24 [1.07–1.44]; p = 0.005 1.27 [1.10–1.48]; p = 0.002 1.26 [1.09–1.46]; p = 0.002
1.27 [0.92–1.74]; p = 0.143 1.27 [0.93–1.73]; p = 0.129 n.a.
2.20 [0.95–1.52]; p = 0.124 1.29 [1.01–1.65]; p = 0.039 n.a.
1.28 [1.00–1.63]; p = 0.050 1.32 [1.03–1.70]; p = 0.028 n.a.
Model 1: adjusted for age and gender; model 2: adjusted for age, gender, body mass index, smoking, and hypertension; model 3: adjusted for the covariates included in model 2, and additionally for type 2 diabetes. Standardized adjusted hazard ratios together with the 95% confidence intervals (CI) are given. Apo A1, apolipoprotein A1; Apo B, apolipoprotein B; LDL, low-density lipoprotein; LDL-PPD, LDL peak particle diameter; HDL, high-density lipoprotein.
Fig. 1. Lipid factors derived from factor analysis as predictors of vascular events. (A) Total study population – model 1: adjusted for age and gender; model 2: adjusted for age, gender, body mass index, smoking, and hypertension; model 3: adjusted for the covariates included in model 2, and additionally for type 2 diabetes. Standardized adjusted hazard ratios [95% confidence intervals] are given. (B) Subgroup analyses with respect to the glycemic state – standardized hazard ratios [95% confidence intervals] are given adjusted for age, gender, body mass index, smoking, and hypertension. LDL, low-density lipoprotein; HDL, high-density lipoprotein; HR, hazard ratio; NFG, normal fasting glucose; IFG, impaired fasting glucose; T2DM, type 2 diabetes.
488
H. Drexel et al. / Atherosclerosis 208 (2010) 484–489
tern strongly predicted future cardiovascular events in patients with manifest T2DM. Pathophyisologically, high triglycerides, low HDL cholesterol, low apolipoprotein A1, and small dense LDL are four different facets of a common metabolic pathway which is driven by insulin resistance [15]. Together, these parameters constitute a lipid pattern which drives accelerated progression of atherosclerotic vascular disease. Because predictive power is lost when only single components from a pattern are considered, the use of an integrative factor should increase discriminative strength. Therefore, it appears adequate to address the comprehensive risk conferred by a pattern of lipid abnormalities by factor analysis, a mathematical technique that integrates common aspects of several individual variables into a single new variable. Indeed, with this novel approach, in our study the HDL-related lipid factor robustly predicted vascular events. The LDL-related factor – which in a similar way integrated common properties of total cholesterol, LDL cholesterol, and apolipoprotein B – in contrast was not associated with vascular events. Importantly, the low HDL cholesterol, high triglycerides, small LDL particle pattern of dyslipidemia was most pronounced and most strongly predicted vascular risk in patients with T2DM. Statins predominantly decrease LDL cholesterol; the risk reduction achieved with statins may therefore be considered to be primarily due to lowering of LDL cholesterol. Because statins only marginally (5–8%) raise HDL cholesterol [16], our data support the notion that the paramount risk indicated by low HDL cholesterol is not adequately addressed by statin therapy. Here the question arises how this risk could be intervened. Life style changes have the potential to improve dyslipidemia through interaction at the underlying disorders of central obesity and insulin resistance, which themselves strongly predict future cardiovascular events as we have shown in a similar population [9,17]. In the light of our data, a focus on more aggressive life style interventions that improve triglycerides, HDL cholesterol, and small dense LDL, e.g. smoking cessation, dietary and exercise counselling appears mandatory. Perhaps the success with drugs such as statins has detracted our attention too much from life style modification of lipids. Among pharmacological interventions, nicotinic acid is a powerful compound to improve triglycerides, HDL cholesterol, small dense LDL, and in addition the independent risk factor lipoprotein(a). There is some evidence from trials that an intervention with nicotinic acid is beneficial in combination with statins [18,19]. The AIM-HIGH [20] and HPS-2 THRIVE [21] trials currently test the potential of a statin-nicotinic acid combination to reduce cardiovascular risk over and above statin monotherapy. Because of the disappointing results of the FIELD study [22], fibrates at present appear less promising, but the results of the fenofibrate arm of the ACCORD trial [23] are still being awaited. Torcetrapib, an inhibitor of CETP that strongly increases HDL cholesterol, did not meet the high expectations [24]. The strengths of our study include four aspects: first, the observational design ascertained that we investigated patients under real-world conditions without a bias from patient selection criteria inherent in interventional studies. Second, CAD was verified by angiography in every single patient. Third, we achieved an almost complete follow-up (99.6%), which is exceptionally high for an observational study. Fourth, we have investigated the power of a broad array of lipids, lipoproteins, and apolipoproteins as predictors of cardiovascular events and integrated them into factor analysis. Taken together, our population is representative, well characterised, and meticulously monitored. As a possible limitation of our investigation, despite coverage of about 2750 patient-years, the sample size warrants consideration. From our data we cannot exclude that, with the greater statistical power of a larger study or a longer follow-up period, an associ-
ation between weaker risk factors and vascular events could be demonstrable. In conclusion, our results suggest that after LDL lowering with statins, HDL raising/triglyceride lowering therapy should be considered as the next important step in lipid intervention for CAD patients, particularly in those with T2DM. Conflict of interest None declared. Acknowledgements The VIVIT Institute thanks Dr. Egmond Frommelt and the Innovationsstiftung of the Liechtenstein Global Trust (LGT) Bank (Bendern, Liechtenstein), Dr. Karl Josef Hier and the Peter Goop Stiftung (Vaduz, Liechtenstein), the Fachhochschule Dornbirn (Dornbirn, Austria), and the Institute for Clinical Chemistry at the Academic Teaching Hospital Feldkirch (Feldkirch, Austria) for providing us with generous research grants. We are grateful to Franz Rauch and the Vorarlberger Industriellenvereinigung (Bregenz, Austria), to Dr. Peter Woess and the Vorarlberger Aerztekammer (Dornbirn, Austria), to Mag. Gabriela Duer and the Vorarlberger Landesregierung, to Dr. Elmar Bechter, Landessanitätsdirektor (Bregenz, Austria) and to Luis Patsch, former Director, Vorarlberger Landeskrankenhaus-Betriebsgesellschaft, for continuously supporting our Research Institute. References [1] McEwen LN, Kim C, Karter AJ, et al. Risk factors for mortality among patients with diabetes: the Translating Research Into Action for Diabetes (TRIAD) study. Diabetes Care 2007;30:1736–41. [2] Cefalu WT. Glycemic targets and cardiovascular disease. N Engl J Med 2008;358:2633–5. [3] Dluhy RG, McMahon GT. Intensive glycemic control in the ACCORD and ADVANCE trials. N Engl J Med 2008;358:2630–3. [4] Baigent C, Keech A, Kearney PM, et al. Efficacy and safety of cholesterollowering treatment: prospective meta-analysis of data from 90,056 participants in 14 randomised trials of statins. Lancet 2005;366:1267–78. [5] Shepherd J, Barter P, Carmena R, et al. Effect of lowering LDL cholesterol substantially below currently recommended levels in patients with coronary heart disease and diabetes: the Treating to New Targets (TNT) study. Diabetes Care 2006;29:1220–6. [6] Castelli WP, Garrison RJ, Wilson PW, et al. Incidence of coronary heart disease and lipoprotein cholesterol levels. The Framingham Study. JAMA 1986;256:2835–8. [7] Barter P, Gotto AM, LaRosa JC, et al. HDL cholesterol, very low levels of LDL cholesterol, and cardiovascular events. N Engl J Med 2007;357:1301–10. [8] Amarenco P, Goldstein LB, Callahan III A, et al. Baseline blood pressure, low- and high-density lipoproteins, and triglycerides and the risk of vascular events in the Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) trial. Atherosclerosis 2009;204:515–20. [9] Saely CH, Aczel S, Marte T, et al. The metabolic syndrome, insulin resistance, and cardiovascular risk in diabetic and nondiabetic patients. J Clin Endocrinol Metab 2005;90:5698–703. [10] Drexel H, Amann FW, Beran J, et al. Plasma triglycerides and three lipoprotein cholesterol fractions are independent predictors of the extent of coronary atherosclerosis. Circulation 1994;90:2230–5. [11] Chobanian AV, Bakris GL, Black HR, et al. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA 2003;289:2560–72. [12] Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1. Diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabet Med 1998;15:539–53. [13] Drexel H, Aczel S, Marte T, et al. Is atherosclerosis in diabetes and impaired fasting glucose driven by elevated LDL cholesterol or by decreased HDL cholesterol? Diabetes Care 2005;28:101–7. [14] Alpert JS, Thygesen K, Antman E, Bassand JP. Myocardial infarction redefined—a consensus document of The Joint European Society of Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction. J Am Coll Cardiol 2000;36:959–69. [15] Taskinen MR. Diabetic dyslipidaemia: from basic research to clinical practice. Diabetologia 2003;46:733–49. [16] Schaefer EJ, Asztalos BF. The effects of statins on high-density lipoproteins. Curr Atheroscler Rep 2006;8:41–9.
H. Drexel et al. / Atherosclerosis 208 (2010) 484–489 [17] Hoefle G, Saely CH, Aczel S, et al. Impact of total and central obesity on vascular mortality in patients undergoing coronary angiography. Int J Obes (Lond) 2005;29:785–91. [18] Brown BG, Zhao XQ, Chait A, et al. Simvastatin and niacin, antioxidant vitamins, or the combination for the prevention of coronary disease. N Engl J Med 2001;345:1583–92. [19] Taylor AJ, Sullenberger LE, Lee HJ, Lee JK, Grace KA. Arterial biology for the investigation of the treatment effects of reducing cholesterol (ARBITER) 2: a double-blind, placebo-controlled study of extended-release niacin on atherosclerosis progression in secondary prevention patients treated with statins. Circulation 2004;110:3512–7. [20] AIM HIGH: Niacin Plus Statin to Prevent Vascular Events. www.clinicaltrials. gov, 2009 (accessed 07.07.09).
489
[21] Treatment of HDL to Reduce the Incidence of Vascular Events HPS2-THRIVE. www.clinicaltrials.gov, 2009 (accessed 07.07.09). [22] Keech A, Simes RJ, Barter P, et al. Effects of long-term fenofibrate therapy on cardiovascular events in 9795 people with type 2 diabetes mellitus (the FIELD study): randomised controlled trial. Lancet 2005;366:1849– 61. [23] Ginsberg HN, Bonds DE, Lovato LC, et al. Evolution of the lipid trial protocol of the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial. Am J Cardiol 2007;99:56i–67i. [24] Tall AR, Yvan-Charvet L, Wang N. The failure of torcetrapib: was it the molecule or the mechanism? Arterioscler Thromb Vasc Biol 2007;27:257–60.