Cholesterol Goal Attainment in Hypertensive Patients:The Impact of Metabolic Syndrome Components

Cassio Jose de Oliveira Rodrigues, M.D.,1 Hercules Ferreira Ribeiro, M.D.,1 Artur Beltrame Ribeiro, Ph.D.,1

Maria Teresa Zanella, Ph.D.,2 and Marcelo Costa Batista, Ph.D.1

Abstract

Background: Hypertension and dyslipidemia are potentially modifiable cardiovascular risk factors.Methods: We studied hypertensive outpatients regarding goal attainment in controlling dyslipidemia, accordingto individual cardiovascular risk profile, following the National Cholesterol Education Program Adult Treat-ment Panel III (NCEP ATP III) guidelines. Factors of goal attainment for low-density lipoprotein cholesterol(LDL-C) were determined.Results: Of the 1,202 patients, this study included 886 (73.8% female, 59.9 – 11.1 years) with available data todetermine cardiovascular risk. Overall, 544 (61.4%) had LDL-C within the goal. Individuals with inappropriateLDL-C were older, had higher systolic blood pressure (SBP), and were more likely to have metabolic syndrome,diabetes, and cardiovascular disease (CVD) and were less likely to show a controlled blood pressure. There wasa progressive worsening of LDL-C control as the number of components of metabolic syndrome increased. Therewas also a progressive increase in the percentage of patients with inappropriate LDL-C with the increase incardiovascular risk. In a logistic regression model including LDL-C inadequacy as a dependent variable, onlyage, diabetes, and CVD were predictors of inappropriate LDL-C. Moreover, even with correction for demo-graphic and clinical variables, the inappropriate LDL-C was an independent predictor of CVD.Conclusions: The control of dyslipidemia in hypertensive patients is far from ideal and results are even worse inindividuals with CVD.

Introduction

Hypertension and dyslipidemia are important modi-fiable risk factors for cardiovascular disease (CVD).1

Several studies have demonstrated that reducing cholesterolwill decrease the number of cardiovascular events.2 TheNational Cholesterol Education Program Adult TreatmentPanel III (NCEP ATP III) guidelines identify low-densitylipoprotein cholesterol (LDL-C) as a causative factor forcoronary heart disease (CHD) and a target for lipid-loweringtherapy.3,4 The guidelines recommend aggressive reductionof LDL-C according to individual baseline risk of futurecardiovascular events, defining LDL-C goals according to thecardiovascular risk.

A number of studies5–14 in different populations haveshown a low attainment of cholesterol goals, concluding thatdyslipidemia is undertreated. The underprescription or dis-continuation of statins and other lipid-lowering medications(LLM) as well as the poor adherence to prescribed treatment

are issues that are related with the low prevalence of patientsachieving all lipid profile goals.15 The aim of this study wasto evaluate the adequacy of lipid profiles as well as deter-minants influencing the control of dyslipidemia among hy-pertensive patients with different cardiovascular factors.

Methods

We included adults treated at the Integrated Center forHypertension and Cardiovascular Metabolism of the FederalUniversity of Sao Paulo (UNIFESP). The medical recordsfor each patient were reviewed, and the most recent dataavailable were considered in the analysis. Only participantsclassified as having hypertension and with a completelipid profile were included. Total cholesterol, high-densitylipoprotein cholesterol (HDL-C), and triglycerides weremeasured enzymatically, and LDL-C and very-low-densitylipoprotein cholesterol (VLDL-C) concentrations were cal-culated by the Friedewald formula.

1Nephrology and 2Endocrinology, Divisions of the Kidney and Hypertension Hospital, Federal University of Sao Paulo, Sao Paulo, Brazil.

METABOLIC SYNDROME AND RELATED DISORDERSVolume 10, Number 3, 2012@ Mary Ann Liebert, Inc.Pp. 195–201DOI: 10.1089/met.2011.0133

195195195

Participants were categorized into NCEP ATP III3 riskcategories based on the 10-year risk of CHD calculated usingthe Framingham risk score (FRS). They were assigned to thefollowing categories: Low risk, 0 or 1 risk factors for CHD;intermediate risk, ‡ 2 risk factors and FRS < 20%; high risk,CHD or CHD risk equivalent (diabetes or CVD) and/orFRS > 20%; and very high risk, diabetes and CVD. Partici-pants were considered to have an appropriate LDL-C iftheir LDL-C concentration was < 160 mg/dL for low risk,< 130 mg/dL for intermediate risk, < 100 mg/dL for highrisk, and < 70 mg/dL for very high risk. Dyslipidemia wasdefined as LDL-C, HDL-C, or non-HDL levels exceeding therisk group–specific NCEP ATP III goal or current use ofLLM, such as hydroxymethylglutaryl-coenzyme A (CoA)reductase inhibitors, bile acid sequestrants, fibric acid deri-vates, or niacin.

Patients were considered to have hypertension if they re-ceived antihypertensive drug treatment or if their bloodpressure was 140/90 mmHg or higher. Blood pressure con-trol was defined according to recommendations of the VIIReport of the Joint National Committee on Prevention, De-tection, Evaluation and Treatment of High Blood Pressure.16

Metabolic syndrome was diagnosed according to NCEPATP III3 and diabetes according to the American DiabetesAssociation criteria.17 Glycosylated hemoglobin (HbA1c)was considered normal if < 7%.

CVD was defined as the presence of CHD (evidence ofsilent myocardial infarction or myocardial ischemia, historyof unstable angina or stable angina pectoris, and history ofcoronary angioplasty or coronary artery surgery) or CHDrisk equivalent (stroke, peripheral arterial disease, abdominalaortic aneurysm, carotid artery disease, renal artery disease),according to American Heart Association guidelines.16 Theglomerular filtration rate (GFR) was estimated using theformulas described in the Modification of Diet in RenalDisease (MDRD) study.18

Data were compared between patients with appropriateand inappropriate LDL-C, and the parameters associatedwith the adequacy of LDL-C were determined. The sameanalysis was performed on a subgroup of diabetics, follow-ing the same criteria.

Data were summarized by the mean – standard deviation(SD) for quantitative traits and by percentage for categor-ical traits. Continuous variables were compared using theStudent t-test and categorical values by the chi-squared test(or Fisher exact test when appropriate). Binary logistic re-gression was employed to identify independent predictorsof inadequate LDL-C, including variables with statisticalsignificance in univariate analysis. The following variableswere included: Age, diabetes, CVD, metabolic syndrome,kidney disease (defined as GFR < 60 mL/min), and inade-quate blood pressure. Other logistic regression modelswere performed to assess the independent association be-tween inadequate LDL-C and CVD. Demographic andclinical variables associated with this outcome in the gen-eral population were entered into the multivariate-adjustedmodels. The multivariate models were performed in stepsand were built for the total study sample as well as forpatients with diabetes. Values of P < 0.05 were consideredto be significant. Analyses were performed using SPSS 17.0software.

The study protocol was approved by the UNIFESP Ethicsin Research Committee. The study was conducted in accor-

dance with the Declaration of Helsinki and with BrazilianNational Ministry of Health Resolution CNS 196/96.

Results

Of 1,202 patients entered in the study, 316 (26.3%) had nodiagnosis of hypertension, no complete lipid panel, or noavailable data to calculate FRS and were thus excluded fromthe final analysis. A total of 886 participants (73.8% female,age 59.9 – 11.1 years, 59% white) were included in the eval-uation.

Baseline characteristics of the population are shown inTable 1. A total of 544 (61.4%) patients showed appropriateLDL-C according to risk stratification. Dyslipidemia wasidentified in 663 (74.8%) participants and 285 (32.2%) weretaking LLM (29.9% on statins and 3.1% on fibrates). Amongpatients with dyslipidemia, 43% were taking LLM (93%of them were on statins) and only 48.4% met their re-commended LDL-C goal. When considering the 285 patientstaking LLM, the percentage of patients showing appropri-ate LDL-C was 60%. Attainment of the recommended non-HDL-C goal was found in 63.1% of all patients, 50.7% ofdyslipidemic subjects, and 62.1% of those taking LLM. Theprevalence of patients achieving all lipid profile goals (LDL-C, HDL-C, and non-HDL-C) was 36.5% among the studypopulation, 15.1% of dyslipidemic subjects, and 35.1% ofthose taking LLM. Overall, only 32.5% had appropriateblood pressure levels and 42.3% had a blood pressure of140/90 mmHg or lower, although 96% were taking antihy-pertensive drugs.

As shown in Table 1, patients not meeting appropriateLDL-C were older and more likely to have diabetes or CVDand to receive three or more antihypertensive drugs, al-though they were less likely to show controlled blood pres-sure. Considering the NCEP ATP III risk categories, 7.9% ofthe participants were at low risk, 40.3% were at intermediaterisk, 44.3% were at high risk, and 7.3% were at very high riskof CHD. As shown in Fig. 1, the percentage of LDL-C andgoal attainment of global lipid profile differed significantlybetween CHD risk categories, with the highest control per-centages among those with lower risk. Conversely, there wasan inverse association between the prevalence of patientstaking LLM and cardiovascular risk stratification. As shownin Fig. 2, there was a lower percentage of attainment of therecommended LDL-C goal as the number of components ofmetabolic syndrome increased.

When considering only patients with diabetes, 156(44.8%) showed appropriate LDL-C; dyslipidemia wasidentified in 83.9%, and 44% were taking LLM. Among di-abetic patients with dyslipidemia, 34.2% showed appropri-ate LDL, and 52.4% were taking LLM. As shown in Table 2,diabetic patients without appropriate LDL-C had highersystolic blood pressure (SBP) and diastolic blood pressure(DBP) levels and were more prone to have CVD and lesslikely to receive LLM and oral antidiabetic drugs. Unlikethat found in the overall population, in patients with dia-betes there was no association between the components ofmetabolic syndrome and the percentage of patients withappropriate LDL-C.

In the binary logistic regression model including LDL-Cgoal attainment as a dependent variable, only age, diabetes,and CVD were found to be predictors of inappropriateLDL-C (Table 3). In all regression models built to assess the

196 RODRIGUES ET AL.

FIG. 1. Percentage of patients with appropriate low-density lipoprotein cholesterol (LDL-C; black), appropriate global lipidprofile (dark gray), and use of lipid-lowering medications (LLM; light gray) according to Framingham Risk Stratification. (*)P < 0.001.

Table 1. Patient Characteristics in the Total Population and in Patients with Versus

without Adequate Low-Density Lipoprotein Cholesterola

Total population (n = 886) Adequate LDL-C (n = 544) Inadequate LDL-C (n = 342) P

Female gender 73.8% 75% 71.9% 0.312Age (years) 59.9 – 11.1 58.9 – 11.6 61.6 – 10.1 < 0.001Race (white) 59% 60.1% 57.2% 0.460BMI (kg/m2) 31.1 – 29.4 30.6 – 10.6 31.8 – 10.7 0.107Systolic blood pressure 138.9 – 21.5 137.7 – 21.1 140.7 – 22.0 0.040Diastolic blood pressure 85.5 – 12.3 85.2 – 12.5 86.0 – 12.0 0.367Diabetes 39.3% 28.7% 56.1% < 0.001Metabolic syndrome 46.0% 41% 54.1% < 0.001Smoking 7.8% 7.7% 7.9% 0.925CVD 15.5% 7.7% 27.8% < 0.001Blood pressure goal 32.5% 35.8% 27.2% 0.007‡ 3 antihypertensive drugs 36.6% 32.7% 42.7% 0.003LLM 32.2% 31.4% 33.3% 0.556Total cholesterol (mg/dL) 188.3 – 39.8 173.9 – 35.1 211.3 – 35.8 < 0.001HDL-C (mg/dL) 51.7 – 13.6 52.5 – 13.5 50.4 – 13.7 0.023LDL-C (mg/dL) 109.9 – 32.1 95.7 – 25.6 132.6 – 28.3 < 0.001VLDL (mg/dL) 27.6 – 13.7 26.4 – 13.7 29.6 – 13.3 0.001Triglycerides (mg/dL)b 2.10 – 0.20 2.08 – 0.20 2.13 – 0.19 < 0.001Non-HDL-C (mg/dL) 137.2 – 35.3 122.1 – 27.7 161.9 – 32.4 < 0.001HDL-C ‡ 60 mg/dL 23.9% 27% 19% 0.006C-reactive protein (mg/dL)b 0.03 – 0.18 0.04 – 0.09 0.03 – 0.02 0.432Kidney diseasec 16.9% 15.1% 19.9% 0.063Microalbuminuria (mg/min)b 0.39 – 0.64 0.32 – 0.58 0.49 – 0.71 0.072

aAdequate LDL-C: < 160 mg/dL for low risk, < 130 mg/dL for intermediate risk, < 100 mg/dL for high risk, and < 70 mg/dL for very highrisk.

bLog transformed.cGlomerular filtration rate < 60 mL/min [Modification of Diet in Renal Disease (MDRD) equation].LDL-C, low-density lipoprotein cholesterol; BMI, body mass index; CVD, cardiovascular disease; LLM, lipid-lowering medication; HDL-C,

high-density lipoprotein cholesterol; VLDL, very-low-density lipoprotein.

DYSLIPIDEMIA IN HYPERTENSIVE PATIENTS 197

FIG. 2. Percentage of patients with appropriate low-density lipoprotein cholesterol (LDL-C; black), appropriate global lipidprofile (dark gray), and use of lipid-lowering medications (light gray) according to metabolic syndrome components. (*)P < 0.001.

Table 2. Patient Characteristics in the Total Diabetic Population and in Patients with Versus

without Adequate Low-Density Lipoprotein Cholesterol

Total population (n = 348) Adequate LDL-C (n = 156) Inadequate LDL-C (n = 192) P

Female gender 69.5% 69.9% 69.3% 0.904Age (years) 61.4 – 10.2 61.3 – 10.3 61.5 – 10.1 0.863Race (white) 49.7% 64.6% 58.7% 0.315BMI (kg/m2) 33.2 – 15 33.2 – 17.4 33.2 – 12.8 0.998Systolic blood pressure 138.6 – 21.2 135.8 – 19.2 140.8 – 22.4 0.029Diastolic blood pressure 83.8 – 12.3 82.3 – 12.5 84.9 – 12.0 0.054Metabolic syndrome 62.9% 62.2% 63.5% 0.794Smoking 6% 5.8% 6.3% 0.851CVD 18.7% 7.1% 28.1% < 0.001Diabetic retinopathy 10.6% 9.0% 12.0% 0.366Blood pressure goal 29.3% 31.4% 27.6% 0.438‡ 3 antihypertensive drugs 39.7% 35.9% 42.7% 0.196LLM 44% 53.2% 36.5% 0.002Insulin 17.5% 20.5% 15.1% 0.187Antidiabetic oral drugs 70.4% 76.3% 65.6% 0.03Glucose(mg/dL) 138.8 – 49.2 134.1 – 44.4 142.6 – 52.6 0.114Glycated hemoglobin < 7% 52.9% 54.1% 51.9% 0.695Total cholesterol (mg/dL) 181.7 – 40.0 157.3 – 32.4 201.5 – 34.3 < 0.001HDL-C (mg/dL) 49.7 – 14.1 48.8 – 11.8 50.4 – 15.6 0.296LDL-C(mg/dL) 102.9 – 31.6 77.9 – 15.5 123.2 – 26.3 < 0.001VLDL (mg/dL) 29.0 – 14.8 28.7 – 16.3 29.3 – 13.5 0.677Triglycerides (mg/dL)a 2.12 – 0.21 2.10 – 0.21 2.13 – 0.20 0.588Non-HDL-C (mg/dL) 131.4 – 35.6 106.6 – 24.0 151.9 – 30.3 < 0.001HDL-C ‡ 60 mg/dL 19.3% 16.7% 21.4% 0.27C-reactive protein (mg/dL)a 0.05 – 0.07 0.04 – 0.06 0.06 – 0.09 0.264Kidney diseaseb 16.7% 13.5% 19.3% 0.148Microalbuminuria (mg/min)a 0.77 – 0.78 0.78 – 0.76 0.77 – 0.79 0.509

aLog transformed.bGlomerular filtration rate < 60 mL/min [Modification of Diet in Renal Disease (MDRD) equation].BMI, body mass index; CVD, cardiovascular disease; LLM, lipid-lowering medication; HDL-C, high-density lipoprotein cholesterol;

LDL-C, low-density lipoprotein cholesterol; VLDL, very-low-density lipoprotein.

198 RODRIGUES ET AL.

independent relationship between LDL-C and CVD, inap-propriate LDL-C, even corrected for confounders and otherclinical variables, was a significant, independent predictor ofCVD within the total study sample as well as among patientswith diabetes, as shown in Table 4.

Discussion

We evaluated a group of hypertensive patients demon-strating a poor control of lipid profile, confirming findingsof several other studies, which show that the control ofdyslipidemia remains far from ideal, regardless the guide-lines used and the type of population studied.5–14 In fact,whereas LDL-C goal attainment has exceeded 80% in somerandomized controlled studies, including selected patientswith high rates of compliance and aggressive treatmentstrategies,19,20 studies related to actual clinical practice haveshown less favorable data and confirm our findings.21

Unlike other studies that have examined general popula-tions8,11,13 or patients with CVD,6,7,9,12 this study specificallyaddressed the hypertensive population and allowed us toestablish an association between lipid-lowering treatmentand the adequacy of antihypertensive therapy. Hypertensionis closely linked to metabolic disorders, and the combinationof hypertension and dyslipidemia represents an important

increment in cardiovascular risk. Moreover, both are poten-tially treatable conditions, and evidenced-based data showthat their control reduces cardiovascular events. In ourstudy, most of the population showed inadequate attainmentof lipid profile goals, and almost 40% of them had LDL-Clevels above the recommended goal. Moreover, less thanone-third of the population was taking LLM, and evenconsidering the individuals diagnosed as dyslipidemics,most of them were not receiving medication aimed at itscorrection. These data are in agreement with other studiesthat had specifically evaluated hypertensive patients inrelation to their goal attainment in controlling dyslipide-mia.22–25 A report of 5,376 patients had already shown thatthe simultaneous occurrence of hypertension and dyslipide-mia leads to a cumulative increase in CVD risk and that alarge percentage of hypertensive subjects show high levels ofcholesterol.22 In addition, a survey of 72,351 hypertensivepatients, including 38,116 with dyslipidemia and 15% takingLLM, also showed a poor control of dyslipidemia.23

In our study, goal attainment in regard to lipid profile wasshown to be far from adequate, and control of blood pressurealso did not show better results. Considering the goal set bythe cardiovascular risk assessment, only 32.5% of the popu-lation studied demonstrated appropriate blood pressure andonly 42.3% had values below 140/90 mmHg. Interestingly,we found an association between the presence of inadequatecontrol of LDL-C and blood pressure. Patients with inap-propriate LDL-C levels also demonstrated a trend towardhigher blood pressure and more commonly needed the useof three or more antihypertensive drugs.

In this study, higher age, diabetes, and CVD were inde-pendent predictors of inadequate control of LDL-C. Fur-thermore, when considering CVD as a dependent variable inmultivariate logistic regression models, inappropriate LDLwas an independent predictor of CVD both in the totalpopulation and in patients with diabetes. Other studies havesuggested,26,27 in fact, that patients with high cardiovascularrisk remain with undertreated dyslipidemia more often thanindividuals with lower risk. These studies suggest that phy-sicians usually consider the absolute numbers from a lipidprofile analysis rather than following the recommendations

Table 3. Predictors of Low-Density Lipoprotein

Cholesterol Inadequacy

ORa 95% CI P value

Age 1.02 1.01–1.03 0.039Diabetes 2.94 2.16–4.00 < 0.001Cardiovascular disease 4.51 2.99–6.80 < 0.001Metabolic syndrome 1.21 0.88–1.66 0.251Blood pressure adequacy 1.31 0.93–1.84 0.120Kidney diseaseb 1.11 0.74–1.66 0.621

aAdjusted for factors in the table.bGlomerular filtration rate < 60 mL/min [Modification of Diet in

Renal Disease (MDRD) equation].OR, odds ratio; CI, confidence interval.

Table 4. Association of Low-Density Lipoprotein Cholesterol Inadequacy with Cardiovascular Disease

Outcome OR 95% CI P value

Total populationUnadjusted 4.60 3.10–6.82 < 0.001Adjusted for age, gender and race 4.09 2.60–6.42 < 0.001Adjusted for age, sex, race, metabolic syndrome 4.05 2.57–6.40 < 0.001Adjusted for age, gender, race, metabolic syndrome, and blood pressure 4.05 2.57–6.41 < 0.001Adjusted for age, gender, race, metabolic syndrome, blood pressure, and diabetes 4.03 2.51–6.45 < 0.001Adjusted for age, gender, race metabolic syndrome, blood pressure,

diabetes, smoking, familial history of CVD, and kidney diseasea3.79 2.35–6.11 < 0.001

DiabeticsUnadjusted 5.16 2.59–10.27 < 0.001Adjusted for age, gender and race 5.24 2.43–11.28 < 0.001Adjusted for age, sex, race, metabolic syndrome 5.23 2.43–11.27 < 0.001Adjusted for age, gender, race, metabolic syndrome, and blood pressure 5.14 2.38–11.10 < 0.001Adjusted for age, gender, race, metabolic syndrome, blood pressure, smoking,

familial history of CVD, and kidney diseasea4.84 2.22–10.55 < 0.001

aGlomerular filtration rate < 60 mL/min [Modification of Diet in Renal Disease (MDRD) equation].OR, odds ratio; CI, confidence interval; CVD, cardiovascular disease.

DYSLIPIDEMIA IN HYPERTENSIVE PATIENTS 199

set by the guidelines, and that they did not take into accountindividual risk. NCEP ATP III recommends that lower LDL-Clevels be obtained in individuals with increased cardiovas-cular risk. In our study, patients with higher risk were lesslikely to achieve LDL-C target, although they were morelikely to receive a lipid-lowering therapy. The recommenda-tion of lower LDL-C target in patients at higher cardiovas-cular risk leads to greater difficulty in obtaining adequatelevels of lipids, demonstrating that it is essential to establish amore aggressive treatment in this subset of subjects.

We believe that these findings have not been previ-ously reported specifically in hypertensive populations,but we expect that similar results would be found. Un-doubtedly alarming is the fact that of subjects classified asvery high risk, only 16.9% demonstrated appropriateLDL-C and less than 10% had overall goal attainment intheir lipid profile.

Patients with diabetes also displayed poor control ofdyslipidemia. Less than half the patients had an appropriateLDL-C, where the percentage among subjects with dyslipi-demia was less than 35%. Again, goal attainment was lowerin subjects with associated CVD, particularly those withmore rigid goals. Diabetic patients with appropriate LDL-Cwere more likely to receive LLM and antidiabetic drugs,suggesting that most usually received aggressive treatmentor had better compliance. Because patients with diabetesoften have multiple co-morbidities, many of them receivingmore than five drugs, we believe that the reduced compli-ance may represent the main factor for the low use of lipid-lowering therapy among them.

We have already demonstrated that the presence of met-abolic syndrome in hypertensive patients is associated withan increased prevalence of CVD, although when consideringthose with diabetes such a relationship was no longerobserved.28 In the present study, individuals without ap-propriate LDL-C were more commonly diagnosed withmetabolic syndrome. Furthermore, the prevalence of con-trolled LDL-C decreased as the number of components of thesyndrome increased. This relationship was not found whenwe analyzed only patients with diabetes. These findingsprovide additional evidence that patients with more riskfactors show the least control of dyslipidemia.

Our study has some limitations. Data were obtained frommedical chart documentation, which may have allowedsome omission of information, especially in relation to car-diovascular risk stratification. However, any additional in-formation would allow classification in high-risk groups, andconsequently more rigid lipid goals, leading to a possiblefinding of a worse control of lipid profile. We have no in-formation about dose titration. The cross-sectional nature ofthe study also does not allow us to infer any causal rela-tionship. We did not assess compliance, which may repre-sent an important factor in relation to poor control of bothdyslipidemia and hypertension. Even considering theselimitations, our findings demonstrate that it is imperative torecognize the risk factors on an individual basis and to getbetter results on lipid-lowering treatment among patientswith hypertension.

Our study demonstrates that the control of dyslipidemiaremains a cornerstone in the treatment of the hypertensivepopulation and its low goal attainment is related to a poorblood pressure control. Inappropriate LDL-C is an indepen-dent and significant predictor of CVD.

Acknowledgment

This work was supported by The National Council forScientific and Technological Development/‘‘CNPQ’’–Brazil.

Author Disclosure Statement

No competing financial interests exist.

References

1. Stamler J, Wentworth D, Neaton JD. Prevalence and prog-nostic significance of hypercholesterolemia in men withhypertension: Prospective data on the primary screenees ofthe Multiple Risk Factor Intervention Trial. Am J Med 1986;80:33–39.

2. Holme I. An analysis of randomized trials evaluating theeffect of cholesterol reduction on total mortality and coro-nary heart disease incidence. Circulation 1990;82:1916–1924.

3. Third Report of the National Cholesterol Education Program(NCEP) Expert Panel on Detection, Evaluation, and Treat-ment of High Blood Cholesterol in Adults (Adult TreatmentPanel III) final report. Circulation 2002;106:3143–3421.

4. Grundy SM, Cleeman JI, Merz CN, et al. Implications ofrecent clinical trials for the National Cholesterol EducationProgram Adult Treatment Panel III guidelines. Circulation2004;110:227–239.

5. EUROASPIRE I and II Group. European Action on Sec-ondary Prevention by Intervention to Reduce Events. Clin-ical reality of coronary prevention guidelines: A comparisonof EUROASPIRE I and II in nine countries. Lancet 2001;357:995–1001.

6. Whincup PH, Emberson JR, Lennon L, et al. Low prevalenceof lipid lowering drug use in older men with establishedcoronary heart disease. Heart 2002;88:25–29.

7. Majumdar SR, Gurwitz JH, Soumerai SB. Undertreatment ofhyperlipidemia in the secondary prevention of coronary ar-tery disease. J Gen Intern Med 1999;14:711–717.

8. Ford ES, Mokdad AH, Giles WH, et al. Serum total choles-terol concentrations and awareness, treatment, and controlof hypercholesterolemia among US adults: Findings from theNational Health and Nutrition Examination Survey, 1999 to2000. Circulation 2003;107:2185–2189.

9. Cohen MV, Byrne MJ, Levine B, et al. Low rate of treatmentof hypercholesterolemia by cardiologists in patients withsuspected and proven coronary artery disease. Circulation1991;83:1294–1304.

10. Saydah SH, Fradkin J, Cowie CC. Poor control of risk factorsfor vascular disease among adults with previously diag-nosed diabetes. JAMA 2004;291:335–342.

11. Pearson TA, Laurora I, Chu H, et al. The lipid treatmentassessment project (L-TAP): A multicenter survey to evalu-ate the percentages of dyslipidemic patients receiving lipid-lowering therapy and achieving low-density lipoproteincholesterol goals. Arch Intern Med 2000;160:459–467.

12. Nag SS, Daniel GW, Bullano MF, et al. LDL-C goal attain-ment among patients newly diagnosed with coronary heartdisease or diabetes in a commercial HMO. J Manag CarePharm 2007;13:652–663.

13. Barham AH, Goff DC, Jr., et al. Appropriateness of choles-terol management in primary care by sex and level of car-diovascular risk. Prev Cardiol 2009;12:95–101.

14. Paragh G, Mark L, Zamolyi K, et at. Lipid-modifying ther-apy and attainment of cholesterol goals in Hungary: Thereturn on expenditure achieved for lipid therapy (REALITY)study. Clin Drug Investig 2007;27:647–660.

200 RODRIGUES ET AL.

15. Daskalopoulou SS, Doonan RJ, Mikhailidis DP. Under-treatment of hypercholesterolemia. Curr Med Res Opin 2010;26:439–443.

16. Chobanian A, Bakris G, Cushman W, et al. The SeventhReport of the Joint National Commitee on Prevention, De-tection, Evaluation, and Treatment of High Blood Pressure.JAMA 2003;289:2560–2572.

17. Executive summary: Standards of medical care in diabetes–2010. Diabetes Care 2010;33(Suppl 1):S4–S10.

18. Levey AS, Bosch JP, Lewis JB, et al. A more accuratemethod to estimate glomerular filtration rate from serumcreatinine: A new prediction equation. Modification of Dietin Renal Disease Study Group. Ann Intern Med 1999,;130:461–470.

19. Davidson M, Ma P, Stein EA, et al. Comparison of effects onlow-density lipoprotein cholesterol and high-density lipo-protein cholesterol with rosuvastatin versus atorvastatinin patients with type IIa or IIb hypercholesterolemia. AmJ Cardiol 2002;89:268–275.

20. Garmendia F, Brown AS, Reiber I, et al. Attaining UnitedStates and European guideline LDL-cholesterol levels withsimvastatin in patients with coronary heart disease (theGOALLS study). Curr Med Res Opin 2000;16:208–219.

21. Krobot KJ, Yin DD, Alemao E, et al. Real-world effectivenessof lipid-lowering therapy in male and female outpatientswith coronary heart disease: Relation to pre-treatment low-density lipoprotein-cholesterol, pre-treatment coronary heartdisease risk, and other factors. Eur J Cardiovasc Prev Rehabil2005;12:37–45.

22. Zanchetti A. Hyperlipidemia in the hypertensive patient. AmJ Med 1994;96:3S–8S.

23. Hendrix KH, Riehle JE, Egan BM. Ethnic, gender, and age-related differences in treatment and control of dyslipidemiain hypertensive patients. Ethn Dis 2005;15:11–16.

24. Assmann G, Schulte H. The Prospective CardiovascularMunster (PROCAM) study: Prevalence of hyperlipidemia inpersons with hypertension and/or diabetes mellitus and therelationship to coronary heart disease. Am Heart J 1988;116:1713–1724.

25. O’Meara JG, Kardia SL, Armon JJ, et al. Ethnic and sex dif-ferences in the prevalence, treatment, and control of dysli-pidemia among hypertensive adults in the GENOA study.Arch Intern Med 2004;164:1313–1318.

26. Pearson TA. The undertreatment of LDL-cholesterol: Ad-dressing the challenge. Int J Cardiol 2000;74:S23–S28.

27. Stacy TA, Egger A. Results of retrospective chart review todetermine improvement in lipid goal attainment in patientstreated by high-volume prescribers of lipid-modifyingdrugs. J Manag Care Pharm 2006;12:745–751.

28. Hirota AH, Rodrigues CJO, Borges RL, et al. Performance oftwo metabolic syndrome definitions in the estimation ofcardiovascular disease among hypertensive patients. J ClinHypertens (Greenwich) 2010;12:588–596.

Address correspondence to:Cassio Jose de Oliveira Rodrigues, M.D.

Federal University of Sao Paulo–Sao Paulo, BrazilRua Leandro Duprett, 365, Vila Clementino

Sao Paulo-SP 04025-011Brazil

E-mail: cjorodrigues@uol.com.br

DYSLIPIDEMIA IN HYPERTENSIVE PATIENTS 201