Contemporary Management of Dyslipidemia in High-Risk Patients: Targets Still Not Met

Transcription

1 The American Journal of Medicine (2006) 119, CLINICAL RESEARCH STUDY Contemporary Management of Dyslipidemia in High-Risk Patients: Targets Still Not Met Andrew T. Yan, MD, a Raymond T. Yan, MD, a Mary Tan, BSc, a Daniel G. Hackam, MD, c Kori L. Leblanc, BScPhm, a Heather Kertland, PharmD, a Jennifer L. Tsang, MD, a Shahin Jaffer, MD, MHSc, d Martin L. Kates, MD, e Lawrence A. Leiter, MD, b David H. Fitchett, MD, a Anatoly Langer, MD, MSc, a Shaun G. Goodman, MD, MSc, a for the Vascular Protection (VP) and Guidelines Oriented Approach to Lipid Lowering (GOALL) Registries Investigators a Canadian Heart Research Centre and Terrence Donnelly Heart Centre, Division of Cardiology, and b Division of Endocrinology and Metabolism, St. Michael s Hospital, c Division of Clinical Pharmacology, University of Toronto, Toronto, Ontario, Canada; d Delta, British Columbia, Canada; e Queensway Professional Centre, Etobicoke, Ontario, Canada. ABSTRACT PURPOSE: Our objective was to evaluate treatment patterns and the attainment of current National Cholesterol Education Program (NCEP)-recommended lipid targets in unselected high-risk ambulatory patients. METHODS: Between December 2001 and December 2004, the prospective Vascular Protection and Guidelines Oriented Approach to Lipid Lowering Registries recruited 8056 outpatients with diabetes, established cardiovascular disease (CVD), or both, who had a complete lipid profile measured within 6 months before enrollment. The primary outcome measure was treatment success, defined as the achievement of LDL-cholesterol 2.6 mmol/l (100 mg/dl) according to NCEP guidelines. We examined patient characteristics and use of lipid-modifying therapy in relation to treatment outcome, which included the recently proposed optional LDL-cholesterol target ( 1.8 mmol/l [70mg/dL]) for very high-risk patients. RESULTS: Overall, 78.2% of patients were treated with a statin and 51.2% had achieved the recommended LDL-cholesterol target. Treatment success rate was highest in diabetic patients with CVD (59.6%), followed by nondiabetic patients with CVD (51.8%), and lowest (44.8%) in diabetic patients without CVD (P.0001). Compared with untreated patients, those on statins were more likely to achieve target (34.4% vs 55.9%, P.0001). Of the patients who failed to meet target, only 9.9% were taking high-dose statin, while 29.3% were not prescribed any statin therapy. Among very high-risk patients, 20.8% attained the optional LDL-cholesterol goal. In multivariable analysis, advanced age, male sex, diabetes, coronary artery disease, coronary revascularization, and use of statin were associated with treatment success (all P.0001). CONCLUSION: Despite the well-established benefits of available lipid-modifying drugs, current management of dyslipidemia continues to be suboptimal, with a substantial proportion of patients failing to achieve guideline-recommended lipid targets. There remains an important opportunity to improve the quality of care for these high-risk patients Elsevier Inc. All rights reserved. KEYWORDS: Cardiovascular disease; Diabetes; Dyslipidemia; Treatment; Guidelines Supported by the Canadian Heart Research Centre (a federally incorporated not-for-profit academic research organization), Astra-Zeneca (GOALL), Sanofi Aventis (VP) and Pfizer (VP). The industry sponsors had no involvement in the study conception or design; collection, analysis, and interpretation of data; in the writing of the manuscript; and in the decision to submit the manuscript for publication. Dr. Andrew Yan is supported by the Canadian Institutes of Health Research Fellowship Award, the Canadian Heart Research Centre Fellowship, and the Detweiler Travelling Fellowship. Address reprint requests to Shaun G. Goodman, MD, MSc, Division of Cardiology, St. Michael s Hospital, 30 Bond Street, Room Queen, Toronto, Ontario M5B 1W8, Canada. address: goodmans@smh.toronto.on.ca /$ -see front matter 2006 Elsevier Inc. All rights reserved. doi: /j.amjmed

2 Yan et al Management of Dislipidemia in High-Risk Patients 677 Overwhelming evidence from laboratory investigations and epidemiologic studies supports a causal link between dyslipidemia and atherosclerotic diseases, which constitute the leading cause of death in the developed world. 1 Landmark clinical trials conducted over the past decade also have convincingly demonstrated an important reduction in cardiovascular morbidity and mortality with LDL cholesterol (LDL-C)-lowering therapy. 1 Furthermore, the benefits of treatment appear more pronounced among patients with established or at high risk of developing cardiovascular disease. Therefore, the National Cholesterol Education Program (NCEP) Adult Treatment Panel III (ATP III) guidelines, revised in 2001, continue to focus on LDL-C as the primary therapeutic target and recommend individualized treatment goals tailored to the estimated cardiovascular risk. 2 Similar evidence-based practice guidelines on cholesterol management also have been published in Canada and Europe. 3,4 Despite the well-established efficacy of lipid-modifying therapy, particularly 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase inhibitors (statins), previous studies in the 1990s documented significant gaps between CLINICAL SIGNIFICANCE evidence-based medicine and real world clinical practice For example, among 4888 outpatients enrolled in the Lipid Treatment Assessment Project, overall, only 38% attained the LDL-C goal recommended by NCEP ATP II. In the highest risk group with known coronary artery disease, the proportion of patients achieving target levels was lower at 18%. 5 Yet there exist only limited data on the contemporary treatment of dyslipidemia, 12 especially among high-risk patients. It is critical to re-evaluate current treatment patterns for several reasons. First, the impact of quality improvement strategies and public health initiatives in the past few years on cholesterol management has not been determined. 1 Second, although the ATP III guidelines recognize that diabetes confers a risk equivalent to established cardiovascular disease and set the same LDL-C goal for diabetic patients, 2 it is unknown whether any treatment disparity persists. Third, in view of major randomized clinical trials supporting more aggressive LDL-C lowering, the NCEP has recently proposed a lower LDL-C target of 1.8 mmol/l (70 mg/dl) as a reasonable clinical strategy for very high-risk patients. 13 The implications and feasibility of this intensive In this prospective study of 8056 ambulatory patients at high cardiovascular risk, only about half attained the current guideline-recommended LDL-cholesterol target of 2.6 mmol/l. The majority of patients who failed to achieve this recommended lipid level were either prescribed submaximal doses of statin or not treated with a statin at all. Only about one-fifth of very high risk patients attained the optional LDL-cholesterol target of 1.8 mmol/l. Advanced age, male gender, diabetes, presence of coronary artery disease, previous coronary revascularization, and statin therapy were independently associated with achievement of optimal lipid targets. strategy remain to be elucidated. Finally, it is unclear whether the advent of new and more potent statins has been accompanied by an increase in the proportion of patients reaching the target lipid levels in the real world. 14,15 Accordingly, we sought to determine: the proportion of high-risk patients who achieve the current NCEP target LDL-C levels ( 2.6 mmol/l [100 mg/dl]) in contemporary practice; the proportion of very high-risk patients (with coexisting diabetes and established cardiovascular disease) who reach the recently proposed optional intensive target (LDL-C 1.8 mmol/l [70 mg/dl]); treatment patterns and their relationships with the targets achieved; and the clinical factors associated with the attainment of optimal treatment goals. METHODS Study Design The present study is a cross-sectional analysis of the prospective, observational, Vascular Protection (VP) and Guidelines Oriented Approach to Lipid Lowering (GOALL) Registries. The objectives of these national registries were to examine clinical management practices and to identify gaps between patient care recommended in guidelines and that delivered in the real world. We recruited physicians across Canada through direct mail or fax campaigns, scientific meetings, continuing medical education events, and investigator meetings in previous or other ongoing registries organized by the Canadian Heart Research Centre. Physicians were invited to participate without regard to their prescribing patterns. In total, 278 physicians participated in the VP Registry and 254 participated in the GOALL Registry ( 95% general practitioners); the respective enrollment periods were December 2001 to November 2004 and December 2002 to December Planned 6-month to 1-year follow-up is currently being conducted. Two independent ethics review boards approved the study protocols. Participation in the registries was completely voluntary, and all enrolled patients provided written informed consent. Patient Population Both registries aimed to enroll patients at high risk for or with established cardiovascular disease, and employed similar inclusion criteria, including: coronary artery disease,

3 678 The American Journal of Medicine, Vol 119, No 8, August 2006 and/or peripheral arterial disease, and/or cerebrovascular disease, according to standard definitions. 16 Coronary artery disease was defined as prior coronary artery bypass surgery or percutaneous coronary intervention, previous documentation of myocardial infarction or unstable angina, or stable angina with positive stress test or 50% stenosis of at least one major coronary artery on angiography. Presence of peripheral arterial disease was determined by a history of intermittent claudication, decreased pulses or bruit with ankle-brachial index 0.90, or abnormal duplex ultrasound findings ( 50% stenosis in 1 major artery). Cerebrovascular disease included previous stroke or transient ischemic attack. In the VP Registry, diabetic patients with at least one other cardiovascular risk factor (systolic blood pressure 140 mm Hg, diastolic blood pressure 90 mm Hg, or use of antihypertensive medication; total cholesterol 5.2 mmol/l, or HDL cholesterol 0.9 mmol/l; current cigarette smoking; microalbuminuria) were eligible. 16 The GOALL Registry also included any diabetic patients and elderly patients (age 65 years) with 2 or more risk factors (as above). The presence of diabetes was ascertained based on the current standard diagnostic criteria, previous diagnosis of diabetes by a physician, or the use of anti-hyperglycemic medications or insulin. 16 To enhance the generalizability of the registry findings, there were no specific exclusion criteria, and participating physicians were instructed to enroll consecutive eligible patients. The present study focused on 9525 patients (4926 and 4499 patients enrolled in the VP and GOALL Registries, respectively) with diabetes and/or established cardiovascular disease (defined as the presence of any of the following: coronary artery disease, peripheral arterial disease, or cerebrovascular disease). Data Collection For each patient enrolled, the participating physician reviewed the medical record and abstracted data on a standardized case report form. Data were collected on patient demographics, cardiovascular risk factors, past history of atherosclerotic diseases (as detailed above), medication use, height, weight, heart rate, blood pressure, and routine bloodwork. Lipid-modifying medications encompassed the following drug classes: statins, ezetimibe, fibrates, niacin, and bile acid sequestrants. The doses of the medications also were recorded. Plasma lipid measurements were performed in ordinary commercial laboratories, as in routine clinical practice. A complete lipid profile obtained within 6 months of study enrollment and defined as measurement of fasting total cholesterol (TC), LDL-C, HDL-cholesterol (HDL-C) and triglycerides (TG) on the same day, was available in 8247 patients (87.5%). Patients with familial hypercholesterolemia (TC 9.4 mmol/l, LDL-C 6.8 mmol/l), 17 and/or TG level 4.5 mmol/l (400 mg/dl), which precluded accurate calculation of LDL-C level by the Friedewald formula, were excluded in the present analysis (n 191 [2.3%]). Thus, the study population comprised 8056 patients. Completed case report forms were returned to the Canadian Heart Research Centre and scanned into an electronic database (Teleform TM, Version 7.0, Cardiff Software Inc., San Diego, CA). Queries regarding incomplete or unclear data on case report forms were sent to study investigators for completion or correction. Data Analysis We stratified the study population into 3 mutually exclusive groups: diabetic patients with no known cardiovascular disease (Group I); nondiabetic patients with cardiovascular disease (Group II); and diabetic patients with cardiovascular disease (Group III). Both NCEP ATP III and Canadian guidelines recommend a very similar target LDL-C ( 2.6 mmol/l [100 mg/dl] and 2.5 mmol/l [97 mg/dl], respectively) for these high-risk patients (all 3 Groups). 2,3 The primary outcome measure of this study was treatment success, defined as attainment of the recommended LDL-C goal 2.6 mmol/l (100 mg/dl). Based on the NCEP updated recommendations, 13 we determined the proportion of very high-risk patients (Group III) achieving the optional therapeutic target of LDL-C 1.8 mmol/l (70 mg/dl). Because the TC: HDL-C ratio is a powerful predictor of adverse cardiac events, 18,19 we also evaluated the attainment rate of TC:HDL-C ratio 4, as recommended in the Canadian consensus guidelines. 3 To examine the relationship between the dose of statin prescribed and the lipid profile, we categorized the prescribed dose into 3 groups according to previously published LDL-C reduction comparison charts. 14,20 Atorvastatin mg, fluvastatin 80 mg, lovastatin mg, pravastatin 40 mg, simvastatin mg, and rosuvastatin 5-10 mg daily were considered as standard doses (estimated reduction in LDL-C by 30%-40%); doses below these were considered low. Atorvastatin mg, simvastatin 80 mg, and rosuvastatin mg daily were classified as high dose (estimated reduction in LDL cholesterol 45%). In pooled comparisons of the lipid profiles of patients treated with various statins, atorvastatin and rosuvastatin were grouped together as new generation statins, while fluvastatin, lovastatin, pravastatin, and simvastatin were classified as old statins. We divided the study population into 3 groups ( 65 years, years, and 75 years old) in age-related analyses. Continuous variables are summarized as median with interquartile range or mean standard deviation (SD) for normally distributed data, and group comparisons were made using the Kruskal-Wallis test. Discrete variables are presented as percentage and compared by 2 test. We used the Holm test to correct for post hoc multiple pairwise comparisons. Bivariate correlations were assessed by the nonparametric Spearman rank correlation coefficient. We performed multivariable logistic regression analysis to determine the factors associated with treatment success. Based on the results of prior studies 5,12,21 and bivariate analyses,

4 Yan et al Management of Dislipidemia in High-Risk Patients 679 Table 1 Demographic and Clinical Characteristics Group I Group II Group III Diabetic Patients without Cardiovascular Disease Nondiabetic Patients with Cardiovascular Disease Diabetic Patients with Cardiovascular Disease All Patients (n 2836) (n 3252) (n 1968) (n 8056) Age, years* 61 (53-70) 68 (59-75) 68 (60-74) 66 (57-73) Male sex, % Current smoker, % Family history of premature coronary disease, % Hypertension, % Stable angina, % Myocardial infarction, % Prior PCI, % Prior CABG, % Heart failure, % Peripheral arterial disease, % Transient ischemic attack, % Stroke, % Heart rate, beats per min* 73 (70-80) 70 (63-75) 72 (66-76) 72 (66-78) Systolic blood pressure, mm Hg* 130 ( ) 130 ( ) 130 ( ) 130 ( ) Diastolic blood pressure, mmhg* 80 (70-82) 78 (70-80) 76 (70-80) 80 (70-80) BMI* 31.0 ( ) 27.7 ( ) 30.0 ( ) 29.3 ( ) Serum creatinine, mol/l* 81 (69-95) 91 (79-105) 90 (78-107) 87 (75-102) Statin use, % Low dose, % Standard dose, % High dose, % Nonstatin use, % Abbreviations: BMI body mass index; CABG coronary artery bypass graft surgery; PCI percutaneous coronary intervention. *Median (interquartile range). the predictor variables considered in the model were age, sex, diabetes, smoking status, body mass index, presence of coronary artery disease, cerebrovascular disease or peripheral vascular disease, and use of statins and other lipidmodifying drugs. These covariates were entered with backward stepwise elimination (P.10 for removal) using the likelihood ratio test to arrive at the parsimonious model. We calculated adjusted odds ratios (OR) with 95% confidence intervals (CI), and evaluated interaction terms between statin treatment and other patient characteristics in the model. To examine the temporal pattern of treatment success, the year of registry enrollment also was entered as a categorical variable. Model discrimination was assessed by the c-statistic (area under receiver-operating characteristic curve), and calibration by the Hosmer-Lemeshow goodness-of-fit test. We performed statistical analyses using SPSS version 12.0 (SPSS Inc., Chicago, IL), and defined statistical significance at a 2-sided P value of.05. RESULTS Of the 8056 patients in the present analysis, 2836 (35.2%) had diabetes only (Group I), 3252 (40.4%) had established cardiovascular disease and no diabetes (Group II), and 1968 (24.4%) had coexisting diabetes and cardiovascular disease (Group III). Their demographic and clinical characteristics are shown in Table 1. Patients with cardiovascular disease (Groups II and III) were older and had higher serum creatinine compared to those without (Group I), while diabetic patients (Groups I and III) had higher body mass indices than their nondiabetic (Group II) counterparts (all P.0001). Among patients in Group I, the vast majority (96.2%) had at least one other cardiovascular risk factor other than dyslipidemia and diabetes. Use of Lipid-Modifying Therapy Overall, 6433 patients (79.9%) were prescribed at least one lipid-modifying drug, with 6297 (78.2%) and 244 (3.0%) receiving statin and nonstatin therapy, respectively (Table 1). Group I patients were less frequently given statin therapy (P.0001), but there were no significant differences in the use of nonstatin drugs among the 3 groups. The majority of patients (59.5%) were taking a standard dose of statin; 7.8% were taking a low dose, and 10.8% were taking a high dose. Table 2 summarizes the types and doses of statin prescribed. Atorvastatin (n 3307) and fenofibrate (n 185) were the most commonly used statin and nonstatin agents, respectively. The rates of statin use were 77.1%, 81.6%, and 75.4% among patients aged 65, 65-74, and 75 years, respectively (P.0001 for difference). Men were more likely than women to be treated with statins (79.6% vs 75.2%, P.0001). Overall, the prescription rates of statin

5 680 The American Journal of Medicine, Vol 119, No 8, August 2006 Table 2 Types and Doses of Statins Prescribed Statin No. of Patients Dose (mg/day)* Atorvastatin (10, 20) Fluvastatin (20, 40) Lovastatin (20, 40) Pravastatin (20, 40) Simvastatin (20, 40) Rosuvastatin (10, 10) *Median (interquartile range). were 68.8%, 82.5%, and 78.6% for patients enrolled in the period , 2003, and 2004, respectively (P.001). Lipid Profile by Patient Groups and Demographics Table 3 presents the lipid profile by the 3 patient groups. The average TC and LDL-C levels were lowest in Group III and highest in Group I (P.001 for all pairwise comparisons among 3 groups). TG level was lower in Group II compared with Groups I and III (P.001 for both pairwise comparisons) but did not differ significantly between Group I and III. HDL-C levels and TC:HDL-C ratios also demonstrated significant differences in all post hoc pairwise comparisons (all P.01). Age was correlated inversely with TC, TG, and LDL-C levels, and positively with HDL-C level (all P.001). Women had significantly higher TC ( vs mmol/l), TG ( vs mmol/ L), HDL-C ( vs mmol/l), and LDL-C levels ( vs mmol/l) than men (P.0001 for all differences). Compared with patients not on statin therapy, patients taking statins had significantly lower TC ( vs mmol/l), TG ( vs mmol/l), LDL-C levels ( vs mmol/l) and TC:HDL-C ratios ( vs mmol/l) (P.0001 for all differences). In contrast, there was no significant difference in HDL-C levels between the untreated and treated patients ( vs mmol/l, respectively). Achieving Treatment Goals Overall, 51.2% of patients achieved the target LDL-C 2.6 mmol/l (100 mg/dl), 50.3% had TC:HDL-C ratio 4, and 36.3% attained both optimal LDL-C and TC:HDL-C ratio. The Figure depicts the percentages of patients achieving these goals in the 3 groups. The proportion of patients reaching the LDL-C target was highest in Group III (59.6%), intermediate in Group II (51.8%), and lowest in Group I (44.8%) (P.0001 for 3-group and all pairwise comparisons). In contrast, patients in Group II (55%) were more likely to achieve the ideal TC:HDL-C ratio than Group I (46%) or III (49%) (P.001 for both pairwise comparisons). Compared with patients not prescribed statins, those on therapy more frequently attained the target LDL-C level (34.4% vs 55.9%, P.0001) and TC:HDL-C ratio (39.2% vs 53.5%, P.0001). However, there were no significant differences in the target LDL-C attainment rates among patients taking old versus new statins (54.8% vs 56.4%), and across the various dosages (low dose 52.2% vs standard dose 56.5% vs high dose 55.2%). Among the 3929 patients who did not achieve primary treatment target (LDL-C 2.6 mmol/l [100 mg/dl]), the lipid profiles were as follows: TC mmol/l, LDL-C mmol/l, HDL-C mmol/l, TG mmol/l, and TC:HDL-C ratio Only 9.9% were taking high-dose statin; 53.1% and 7.6% were maintained on standard and low-dose statins, respectively; 29.3% (n 1152) were not prescribed any statin therapy. In the GOALL Registry, physicians were asked to provide the reasons for not prescribing statin for these patients (n 462). The results are summarized in Table 4. Of the patients in Group III (very high risk), the majority (n 1559, or 79.2%) did not achieve the optional LDL-C target of 1.8 mmol/l (70 mg/dl). Among these patients, 12.2% were given a high-dose statin, 66.4% were taking a low or standard dose, and 21.3% were not on any statin. Table 3 Lipid Profile by Patient Groups Group I Group II Group III Diabetic Patients without Nondiabetic Patients with Diabetic Patients with All Patients Cardiovascular Disease Cardiovascular Disease Cardiovascular Disease (n 8056) (n 2836) (n 3252) (n 1968) TC (mmol/l) LDL-C (mmol/l) HDL-C (mmol/l) TG (mmol/l) TC:HDL-C ratio All values are presented as mean SD. To convert cholesterol and triglycerides units from mmol/l to mg/dl, multiply by and 88.57, respectively. Abbreviations: TC total cholesterol; TG triglycerides. P Value (3-group comparison)

6 Yan et al Management of Dislipidemia in High-Risk Patients 681 Figure Achievement of treatment targets by patient groups. CVD cardiovascular disease; TC total cholesterol. Multivariable Analysis In multivariable logistic regression analysis, several demographic and clinical features were independently associated with treatment success (Table 5). The type of statin prescribed (new vs old), use of nonstatin drugs, body mass index, history of cerebrovascular or peripheral vascular disease were not related to treatment success. Because we found no significant interaction effect between statin use and other variables on treatment outcome, the interaction terms were excluded in the final model. The model c- statistic was 0.64 (P.0001), and the P value for Hosmer- Lemeshow goodness-of-fit test was 0.82, indicating that the model provided adequate fit with the data. Compared with patients enrolled in , those enrolled in 2004 were significantly more likely to attain treatment targets (adjusted odds ratio 1.62, 95% confidence interval 1.43 to 1.83, P.001). DISCUSSION In this prospective, observational study of 8056 high-risk ambulatory patients, we demonstrated that only about half achieved the target LDL-C levels or TC:HDL-C ratio recommended in current guidelines. Furthermore, a considerable proportion of patients who failed to reach target were either prescribed submaximal doses of statins, or not treated Table 4 Reasons for Not Prescribing Statin among 462 Patients Who Did Not Achieve Treatment Target (LDL-C 2.6 mmol/l [100 mg/dl]) in the GOALL Registry Reason* % Not high risk or not supported by evidence 7.4 Allergy/Intolerance 5.4 Myopathy 2.4 Elevated liver enzymes 1.5 Nonadherence 18.4 Other 16.9 Not provided 51.3 *Not mutually exclusive. with statins at all. Our study furnishes new insights into the contemporary management of dyslipidemia in high-risk patients, and underscores a substantial opportunity to reduce morbidity and mortality through more aggressive treatment of this potent yet modifiable risk factor. Over the past decade, rapid advances both in our understanding of dyslipidemia and in the available therapeutic armamentarium have necessitated several revisions of the NCEP guidelines. 1 Recognizing that diabetes confers a cardiovascular risk equivalent to that of established atherosclerotic disease, the ATP III set the same LDL-C target ( 2.6 mmol/l [100 mg/dl]) for diabetic patients in In support of this decision, more recent clinical trials have confirmed the benefits of LDL-C lowering as an effective primary prevention strategy for diabetic patients. 22,23 However, in a study of coronary artery disease patients conducted between 1996 and 1998, those with diabetes were in fact less likely to have a lipid profile measured and to receive lipid-modifying medications than their nondiabetic counterparts. 11 Moreover, the impact of ATP III guidelines on the management of diabetic patients with no known cardiovascular disease has not been well studied. 12 In the present analysis, despite the absence of pre-existing cardio- Table 5 Multivariable Analysis: Factors Associated with Treatment Success (LDL-C 2.6 mmol/l [100 mg/dl]) Variable Adjusted Odds Ratios (95% CI) P Value Age years Referent group years 1.30 (1.17 to 1.44) years 1.47 (1.30 to 1.66).0001 Female sex 0.81 (0.74 to 0.89).0001 Diabetes 1.41 (1.27 to 1.57).0001 Coronary artery disease 1.34 (1.19 to 1.52).0001 Prior PCI or CABG 1.50 (1.32 to 1.70).0001 Statin therapy 2.25 (2.01 to 2.52).0001 Abbreviations: CABG coronary artery bypass graft surgery; PCI percutaneous coronary intervention.

7 682 The American Journal of Medicine, Vol 119, No 8, August 2006 vascular disease among patients in Group I, the majority had multiple cardiovascular risk factors and constituted a truly high-risk group. Therefore, the less frequent use of statins and attainment of treatment goals among these patients strongly suggest that undertreatment remains a serious concern and that physicians need to better appreciate the benefits of primary preventive therapy for diabetic patients. Conversely, the importance of diabetes in patients with established cardiovascular disease appears to be well recognized, as reflected by the highest proportion of patients achieving target LDL-C in this group (Group III). Overall, we found that only 51.2% of high-risk patients achieved the recommended target LDL-C level. Our results concur with those of a recent US national survey that documented a LDL-C goal attainment rate of 57% among 2708 high-risk patients. 12 Although these figures represent an encouraging increase compared with the treatment success rate of 18% among 1460 patients with coronary artery disease in the Lipid Treatment Assessment Project, 5 almost half of our patients still failed to obtain the full treatment benefits, with a high average LDL-C of 3.42 mmol/l. Similarly, only 50.3% attained the optimal TC:HDL-C ratio recommended in Canadian guidelines, and even fewer (36.3%) met the targets for both LDL-C and TC:HDL-C ratio. Thus, strategies to improve management of dyslipidemia in high-risk patients are necessary. Several reasons may account for the poor adherence to treatment guidelines. However, lack of awareness of treatment guidelines or knowledge gap does not appear to be the most important contributing factor to the suboptimal management of dyslipidemia in these patients. In the GOALL Registry, physicians correctly identified the target lipid levels (LDL-C 2.6 mmol/l [100 mg/dl] and TC:HDL-C ratio 4) for the majority (96%) of high-risk patients (data not shown). Instead, our results suggest that there are significant barriers to more widespread and successful implementation of current guidelines. These may include inappropriate drug or dose selection, failure to initiate or titrate therapy, patient nonadherence, or limited drug efficacy. Of note, only a minority of untreated patients who were not achieving treatment targets had documented drug intolerance or side effects (Table 4). The observation that similar proportions of patients achieved target LDL-C regardless of the statin dose implies some titration of therapy had probably occurred. Nevertheless, the substantial proportion of treatment failures not on high-dose statin (90.1%) highlights further opportunities to improve care. Importantly, we did not find a significantly higher treatment success rate associated with the use of new compared with old statins. This, at least in part, reflects the infrequent use of high dose statins. While it should be emphasized that our objective was not to compare the efficacy of LDL-C reduction with old versus new statins, we contend that the full potential of newer statins has yet to be realized and that careful monitoring and titration of therapy will remain the cornerstone of treatment success. In July 2004, the NCEP recommended LDL-C 1.8 mmol/l (70 mg/dl) as reasonable clinical strategy for the very high-risk diabetic patients with cardiovascular disease. 13 The Vermont Diabetes Information System trial demonstrated that only 15.7% of 191 very high-risk patients had LDL-C 1.8 mmol/l (70 mg/dl) in 2003, 24 which was comparable with the rate of 17.8% (of 1447 patients) recently reported in NEPTUNE II. 12 Our results confirm that only a minority of such patients (20.8% of 1968 patients in Group III) would have achieved this target in 2001 to 2004, and more effective therapeutic strategies are warranted should this optional target gain more widespread acceptance. As recent clinical trials continue to support more aggressive LDL-C lowering, our data also should serve as a useful benchmark to re-evaluate the attainment of this optional LDL-C goal in the future. The findings that older age and male sex were independent predictors of treatment success are consistent with other studies. 5,12,21 In addition to a possible survivor effect, lipid levels may decrease with increasing age. The previously documented sex disparities in the management of dyslipidemia seem to persist in the current era. 5,7 In a recent study of high-risk women in a managed care setting, Mosca et al showed that only 12% achieved the American Heart Association optimal combined lipid levels. 21 Our study reaffirms the need to narrow this treatment gap. Finally, a history of coronary revascularization might be associated with in-hospital initiation of lipidmodifying drugs, ongoing specialist care, better medication adherence, or more intensive lifestyle interventions that led to greater treatment success. Several limitations should be considered in the interpretation of this study. We did not collect data on duration of statin therapy, dose titration, and detailed dietary information, although dietary and other lifestyle interventions had likely been implemented concurrently according to current guidelines. 2,3 While our study attempted to minimize bias by enrolling consecutive eligible patients in many different practice settings, the nonrandom selection of participating physicians and consenting patients with complete lipid profile might limit the generalizability of our findings. We could not determine the impact of this selection bias against less motivated physicians and patients, who might be even less likely to follow treatment guidelines and attain treatment targets. Nevertheless, our results were probably conservative and underestimated the prevalence of treatment failure in the general population. Lipid measurements were not performed in a central core laboratory. However, this reflects the real world situation where physicians initiate or titrate therapy based on the test results available in their routine practice. Finally, adherence was not assessed by pill count or pharmacy record, although the seldom prescription of high-dose statins (given their safety profile) even among patients not meeting the recommended target strongly suggests that undertreatment by physicians plays a key role.

8 Yan et al Management of Dislipidemia in High-Risk Patients 683 In conclusion, current management of dyslipidemia remains suboptimal, with a persistent under-utilization of effective lipid-modifying therapy, and almost half of the highrisk patients failing to achieve guideline-recommended lipid targets. These findings emphasize the need to titrate therapy and to optimize compliance in order to bridge the gap between evidence-based medicine and current practice, thereby translating the important benefits demonstrated in clinical trials to the real world. ACKNOWLEDGMENTS We thank Sue Francis for her secretarial assistance. We are indebted to all the study investigators and patients who participated in the GOALL and VP Registries. References 1. National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III): final report. Circulation. 2002;106: Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA. 2001;285: Fodor JG, Frohlich JJ, Genest JJ Jr, McPherson PR. Recommendations for the management and treatment of dyslipidemia. Report of the Working Group on Hypercholesterolemia and Other Dyslipidemias. CMAJ. 2000;162: De Backer G, Ambrosioni E, Borch-Johnsen J, et al. European guidelines on cardiovascular disease prevention in clinical practice. Third Joint Task Force of European and Other Societies on Cardiovascular Disease Prevention in Clinical Practice. Eur Heart J. 2003;24: Pearson TA, Laurora I, Chu H, Kafonek S. The lipid treatment assessment project (L-TAP): a multicenter survey to evaluate the percentages of dyslipidemic patients receiving lipid-lowering therapy and achieving low-density lipoprotein cholesterol goals. Arch Intern Med. 2000; 160: EUROASPIRE I and II Group. Clinical reality of coronary prevention guidelines: a comparison of EUROASPIRE I and II in nine countries. EUROASPIRE I and II Group. European Action on Secondary Prevention by Intervention to Reduce Events. Lancet. 2001;357: Svilaas A, Risberg K, Thoresen M, Ose L. Lipid treatment goals achieved in patients treated with statin drugs in Norwegian general practice. Am J Cardiol. 2000;86: Massing MW, Sueta CA, Chowdhury M, et al. Lipid management among coronary artery disease patients with diabetes mellitus or advanced age. Am J Cardiol. 2001;87: Yarzebski J, Spencer F, Goldberg RJ, et al. Temporal trends ( ) in cholesterol level assessment and management practices in patients with acute myocardial infarction: a population-based perspective. Arch Intern Med. 2001;161: Marcelino JJ, Feingold KR. Inadequate treatment with HMG-CoA reductase inhibitors by health care providers. Am J Med. 1996;100: Massing MW, Foley KA, Sueta CA, et al. Trends in lipid management among patients with coronary artery disease: has diabetes received the attention it deserves? Diabetes Care. 2003;26: Davidson MH, Maki KC, Pearson TA, et al. Results of the National Cholesterol Education Program (NCEP) evaluation project utilizing novel e-technology (NEPTUNE) II survey and implications for treatment under the recent NCEP writing group recommendations. Am J Cardiol. 2005;96: Grundy SM, Cleeman JI, Merz CN, et al. Implications of recent clinical trials for the National Cholesterol Education Program Adult Treatment Panel III guidelines. Circulation. 2004;110: Jones PH, Davidson MH, Stein EA, et al. Comparison of the efficacy and safety of rosuvastatin versus atorvastatin, simvastatin, and pravastatin across doses (STELLAR Trial). Am J Cardiol. 2003;92: Schuster H, Barter PJ, Stender S, et al. Effects of switching statins on achievement of lipid goals: Measuring Effective Reductions in Cholesterol Using Rosuvastatin Therapy (MERCURY I) study. Am Heart J. 2004;147: Hackam DG, Tan MK, Honos GN, et al. How does the prognosis of diabetes compare with that of established vascular disease? Insights from the Canadian Vascular Protection (VP) Registry. Am Heart J. 2004;148: Williams RR, Hunt SC, Schumacher MC, et al. Diagnosing heterozygous familial hypercholesterolemia using new practical criteria validated by molecular genetics. Am J Cardiol. 1993;72: Kinosian B, Glick H, Garland G. Cholesterol and coronary heart disease: predicting risks by levels and ratios. Ann Intern Med. 1994; 121: Pedersen TR, Olsson AG, Faergeman O, et al. Lipoprotein changes and reduction in the incidence of major coronary heart disease events in the Scandinavian Simvastatin Survival Study (4S). Circulation. 1998;97: Maron DJ, Fazio S, Linton MF. Current perspectives on statins. Circulation. 2000;101: Mosca L, Merz NB, Blumenthal RS, et al. Opportunity for intervention to achieve American Heart Association guidelines for optimal lipid levels in high-risk women in a managed care setting. Circulation. 2005;111: Collins R, Armitage J, Parish S, et al; Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of cholesterol-lowering with simvastatin in 5963 people with diabetes: a randomised placebo-controlled trial. Lancet. 2003;361: Colhoun HM, Betteridge DJ, Durrington PN, et al. Primary prevention of cardiovascular disease with atorvastatin in type 2 diabetes in the Collaborative Atorvastatin Diabetes Study (CARDS): multicentre randomized placebo-controlled trial. Lancet. 2004;364: Kennedy AG, MacLean CD, Littenberg B, et al. The challenge of achieving national cholesterol goals in patients with diabetes. Diabetes Care. 2005;28: Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomised placebo-controlled trial. Lancet. 2002;360: Cannon CP, Braunwald E, McCabe CH, et al. Intensive versus moderate lipid lowering with statins after acute coronary syndromes. N Engl J Med. 2004;350: LaRosa JC, Grundy SM, Waters DD, et al. Intensive lipid lowering with atorvastatin in patients with stable coronary disease. N Engl J Med. 2005;352: