Cholesterol level goal attainment with statins: clinical management guideline recommendations versus management in actual clinical practice.

STUDY OBJECTIVES: To compare the cholesterol level goal attainment rates in patients receiving simvastatin doses recommended in clinical practice guidelines and simvastatin doses most frequently prescribed in clinical practice versus other statins at various dose levels, and to assess statin adherence rates in patients receiving all statins. DESIGN: Retrospective cohort study. DATA SOURCE: PHARMO database, which contains linked prescription drug information, hospitalization records, and laboratory test results of over 1 million patients in the Netherlands. PATIENTS: A total of 7355 new statin users with available cholesterol level measurements before and 12 months after starting statin treatment between 1999 and 2006. MEASUREMENTS AND MAIN RESULTS: Simvastatin was chosen as the reference drug because policy makers in the Netherlands have promoted the use of generically available statins to reduce costs. Cholesterol level goal attainment rates were compared in patients receiving simvastatin 40 mg/day, which was the statin dose promoted in the 2006 Dutch cardiovascular risk management guidelines, or simvastatin 20 mg/day, which was the most frequently prescribed dose up to 2006, versus other statins at various dose levels. Relative risks (RRs) were adjusted for age, sex, year of therapy initiation, cardiovascular disease, type 2 diabetes mellitus, hypertension, baseline low-density lipoprotein cholesterol level, and adherence during the 3 months before the 12-month follow-up cholesterol measurement. Compared with simvastatin 40 mg/day, cholesterol goal attainment rates were significantly higher with atorvastatin 40 mg/day (RR 1.15, 95% confidence interval [CI] 1.04-1.28) and rosuvastatin 10 mg/day (RR 1.13, 95% CI 1.04-1.23), were similar with atorvastatin 20 mg/day (RR 1.06, 95% CI 0.97-1.16) and rosuvastatin 20 mg/day (RR 1.14, 95% CI 0.93-1.39), and were significantly lower with all other frequently used statin dose levels. Compared with simvastatin 20 mg/day, cholesterol goal attainment was significantly higher with any dose of atorvastatin and rosuvastatin, but were lower with any dose of pravastatin. Goal attainment rates were similar among patients with lower and higher cardiovascular risk. Among the 13-18% of patients who had follow-up cholesterol level measurements at 12 months in all statin groups, the proportion of adherent patients was approximately 75%; this was higher than the proportion of adherent patients in the total population (48-55%), which included patients without follow-up cholesterol levels. CONCLUSION: A larger proportion of patients reached cholesterol lipid goals with simvastatin 40 mg/day. Cholesterol level goals were achieved by many patients using the recommended simvastatin 40 mg/day, but by fewer patients among those using the more commonly prescribed simvastatin 20 mg/day. Therefore, especially in high-risk patients, the choice of statin should be based on baseline cholesterol levels and expected reductions in these levels, and treatment should be adapted if targets are not met. Improved cholesterol level monitoring may increase adherence and cholesterol management.

Determination of non-treatment with statins of high risk patients in The Netherlands.

OBJECTIVE: To determine in the Netherlands what proportions of high risk patients with established cardiovascular disease (CVD) or diabetes mellitus type 2 (DM2) who were not treated with statins on 1 January 2007 and which characteristics were associated with non-treatment. METHODS: From the IPCI GP database patients were selected who were registered with a GP on 1 January 2007 who had a history of either CVD (CVD patients), DM2 (diabetics) or both (diabetics with CVD). The proportion of patients using statins around 1 January 2007 was determined. Associations of patient characteristics with non-treatment were quantified (CVD patients and diabetics only). RESULTS: In all, 19 628 CVD patients, 5006 diabetics and 3767 diabetics with CVD were identified. Of these patients 71%, 54% and 45%, respectively did not use statins. These proportions were similar in the subgroups of patients with recent LDL-C measurements. Among these subgroups the vast majority of non-treated patients was eligible for statin treatment (LDL-C >2.5 mmol/l). The proportion of statin-treated patients was larger among diabetics than among CVD patients. Among CVD patients, female gender, age below 40 years, living in a deprived area, a history of CVD of less than 1 year and arrhythmia were significantly associated with non-treatment. Among diabetics, significant associations were: living in a deprived area and specialist visits in the previous year. In 2003, treatment rates among diabetics were lower, but among CVD patients they were similar. This suggests that the higher treatment rates among diabetics compared to CVD patients in 2007 may be the result of disease-management programmes introduced for diabetics in 2004. CONCLUSION: The majority of patients with established CVD or DM2 were not treated with statins on 1 January 2007. Eligibility for statin treatment may have been overestimated due to unavailability of cholesterol levels among many non-treated patients. Implementation of care programmes for CVD patients may increase treatment rates among eligible CVD patients.

The influence of six cardiovascular polymorphisms on a first event of ischemic heart disease is modified by sex and age.

OBJECTIVE: To examine the contribution of six cardiovascular polymorphisms to the occurrence of a first event of ischemic heart disease (IHD) in a primary care population with a high prevalence of hypertension. Furthermore, we specified the data for sex and age. METHODS: In this cross sectional case-control study, patients with a first event of IHD (157) and event-free controls (571) were studied. Both the groups were genotyped for the angiotensin II type 1 receptor (A1166C), angiotensinogen (M235 T), angiotensin converting enzyme (4656rpt), endothelial nitric oxide synthase (E298D), G-protein beta3 subunit (C825 T), and alpha-adducin (G460W) polymorphisms. Univariate and multivariate odds ratios (ORs) were calculated to assess the association between a first ischemic event and these polymorphisms. Sliding mean analyses were performed to show age-specific associations. RESULTS: Multivariate ORs indicated a protective association for the carrier status of the T-allele of AGT, overall [OR = 0.69 (0.34-0.90)] and for males [OR = 0.58 (0.27-0.89)]. Sliding mean analyses showed a continuous protective association with IHD of the T-allele of AGT with increasing age in males, whereas in females an increased risk for IHD was observed with a maximum OR of 1.6 at the age of 56 years. CONCLUSION: In this population the T-allele of the AGT polymorphism is protective for a first event of IHD in males.

Sex-specific effect of the alpha-adducin (G460W) and AGTR1 (A1166C) polymorphism on carotid intima-media thickness.

OBJECTIVE: In a primary care population covering a broad spectrum of cardiovascular risk (HIPPOCRATES project) the relationship between carotid intima-media thickness (CIMT) and six cardiovascular polymorphisms were analyzed in a cross-sectional study. METHODS: CIMT was assessed in 618 participants, who were genotyped for the AGTR1 (A1166C), AGT (M235T), ACE (4656(rpt)), NOS3 (E298D), GNB3 (C825T) and ADD1 (G460W) polymorphisms. Linear regression analyses were performed to assess the relationship between CIMT and the polymorphisms. RESULTS: The study population included 289 men (46.8%) and 329 (53.2%) women of whom 52.3% were treated with cardiovascular medication. CIMT was significantly associated with age, female sex, use of cardiovascular medication, waist circumference and dyslipidemia. After correction for these covariates, multivariate linear regression analyses showed that only in women the C-allele of AGTR1 was associated with a thicker CIMT (P = 0.03). The T-allele of ADD1 was associated with a smaller CIMT in both men and women, but this only reached statistical significance in women (P = 0.03). CONCLUSION: Although the effect of both genetic variants on CIMT was small, this study showed a statistically significant effect of AGTR1 and ADD1 in women. However, our findings should be viewed as hypothesis-generating and require further confirmation in prospective epidemiological primary care studies.

Large impact of antidiabetic drug treatment and hospitalizations on economic burden of diabetes mellitus in The Netherlands during 2000 to 2004.

OBJECTIVE: To estimate the burden of diabetes mellitus (DM) and its complications in The Netherlands. METHODS: The PHARMO Record Linkage System comprised among others linked drug dispensing, hospital and clinical laboratory data from approximately 2.5 million individuals in The Netherlands. Patients with DM (type 1 and type 2) were included in the study cohort from 2000 to 2004 if they used antidiabetic drugs or had HbA1c >or= 6.5 mmol/L or had a hospitalization for DM or a diabetic complication in the measurement year or in the preceding year. Controls, defined as subjects without a diagnosis of DM and/or subjects not prescribed glucose-lowering medication, were 1:1 matched to patients with diabetes, on birth year, zip code, and gender. Complications (hospitalizations and dispensings for cardiovascular disease/eye problems/amputations) were classified into stages. Complications attributed to DM were estimated as complication stages 1 and 2 among patients minus those among controls. Drug costs were extrapolated to The Netherlands by direct standardization. RESULTS: Among the total population in The Netherlands, the prevalence of DM increased from 2.8% in 2000 to 4.0% in 2004. Severe cardiovascular complications attributed to DM increased from 18,000 to 39,000 patients. Per DM patient the cost of direct treatment attributed to DM increased from Euro 974 in 2000 to Euro 1283 in 2004. Per 100 members of the total population, this increase was from Euro 2764 in 2000 to Euro 5140 in 2004. Most of these costs (65% in 2004) were because of hospitalizations. CONCLUSION: Drug treatment, hospitalizations, and cost attributed to diabetes mellitus have almost doubled between 2000 and 2004, but so did the "background" costs in the general population, perhaps because of preventive efforts.

The association between arterial stiffness and the angiotensin II type 1 receptor (A1166C) polymorphism is influenced by the use of cardiovascular medication.

OBJECTIVE: To examine in a population with a high prevalence of hypertension, the association between six cardiovascular polymorphisms, arterial stiffness and medication use. METHODS: In this cross-sectional study (Hypertension: Interaction and Prevalence of POlymorphisms related to Cardiovascular Risk and the Association to Treatment Efficacy Study project), arterial stiffness was assessed by measuring pulse wave velocity (PWV) in 575 patients in one primary care practice. Patients were genotyped for the angiotensin II type 1 receptor [AGTR1 (A1166C)], angiotensinogen (M235T), angiotensin-converting enzyme (4656rpt), endothelial nitric oxide synthase (E298D), G-protein beta3 (C825T), and alpha-adducin (G460W) polymorphisms. Linear regression analyses were performed to assess the association between polymorphisms and PWV. RESULTS: Thirty percent of the patients (273 men, 302 women) had a carotid-femoral pulse wave velocity above 12 m/s and more than 60% of the patients had a carotid-femoral/carotid-radial PWV (CF/CR ratio) above 1. The CF/CR ratio was significantly associated with age, sex, dislipidemia, cardiovascular medication use and pulse pressure. After correction for these covariates, multivariate linear regression analyses showed that the C allele of AGTR1 was associated with a lower CF/CR ratio. This association was significantly influenced by cardiovascular medication use (P = 0.011), and showed a dose-allele effect, the CF/CR ratio decreasing with the number of C alleles (P = 0.04). CONCLUSION: In a primary care population, this study showed an independent protective dose-allele effect for the presence of C alleles of the AGTR1 polymorphism on PWV. This association, which was influenced by the use of cardiovascular medication, needs further investigations to identify the underlying mechanisms.

Obtaining the family history for common, multifactorial diseases by family physicians. A descriptive systematic review.

OBJECTIVE: To summarize the knowledge on the use of comprehensive family history taking for common, multifactorial diseases in primary health care. DESIGN AND DATA SOURCES: Systematic review of MEDLINE (1966-2008), EMBASE (1986-2008) and Cochrane Library. METHODS: Search terms reflected 'primary care', 'family history' and 'genetics'. Included were original studies, published in the English language, from a primary care setting, investigating family history taking for multifactorial disorders. Methodological criteria (design, size, response rate) were not used to exclude papers. Out of 116 potentially eligible papers, 27 papers were selected: nine studies on opinions, eight studies on actual practice, seven studies on family history tools, and three studies on the patient perspective. Two authors independently extracted the data, and consequently discussed and summarized them. Given the heterogeneity of the studies, outcomes were presented in a qualitative way. RESULTS: Among family physicians, the general opinion was that taking a family history is the task of the primary care physician. However, observational studies of consultations and analyses of medical records showed wide variability and low regular updating. There are no family history tools yet, that are sufficiently feasible and reproducible. Patients and doctors may perceive a positive family history differently, which may cause miscommunication. CONCLUSION: There is a need for research into feasible and high quality tools for detailed family history taking for multifactorial disorders.

Genetic risk of atherosclerotic renal artery disease: the candidate gene approach in a renal angiography cohort.

It is largely unknown to what extent genetic abnormalities contribute to the development of atherosclerotic renal artery disease. Among the potential candidate genes, those of the renin-angiotensin system and the endothelial nitric oxide synthase (eNOS) rank high because of their importance in the atherosclerotic process. We investigated the association of polymorphisms in these genes (the angiotensinogen Met235Thr, the angiotensin-converting enzyme insertion/deletion, the angiotensin II type-1 receptor A1166C, and the eNOS Glu298Asp) with the presence or absence of atherosclerotic renovascular disease in 456 consecutive hypertensive patients referred for renal angiography on the suspicion of renovascular hypertension. Nondiseased normotensive (n=200) and hypertensive (n=154) patients from a family practice served as external controls. Renal artery disease was present in 30% of our angiography group. The Asp allele of the eNOS Glu298Asp polymorphism was associated with atherosclerotic renal artery stenosis with an odds ratio of 1.44 (95% confidence interval 1.00 to 2.09) versus hypertensives with angiographically proven patent arteries, of 1.89 (1.24 to 2.87) versus hypertensive family practice controls, and of 2.09 (1.29 to 3.38) versus normotensive family practice controls. However, this allele also differed significantly between patients with patent renal arteries and normotensive and hypertensive controls. No differences were found with respect to the other genetic polymorphisms. We hypothesize that the Asp allele of the Glu298Asp polymorphism may predispose to the development of atherosclerotic lesions but that renal artery involvement depends on other factors, also.

Potential interaction between acenocoumarol and diclofenac, naproxen and ibuprofen and role of CYP2C9 genotype.

NSAIDs are reported to increase the risk of bleeding in coumarin users. The mechanism underlying this risk is inhibition of platelet aggregation, however a pharmacokinetic mechanism resulting in an increased International Normalised Ratio (INR) was proposed in some case reports in warfarin treated patients. In this retrospective cohort study the influence of diclofenac, naproxen and ibuprofen on the INR of outpatients stabilised on acenocoumarol therapy was investigated. We also determined the role of cytochrome P450 2C9 (CYP2C9) polymorphism on coumarin dosage and INR in NSAID users. The study was carried out at the Groningen Outpatient Thrombosis Service. A retrospective cohort study among patients who received both acenocoumarol and one of the NSAIDs under study was performed. Patients whose INR rose above the upper level of the therapeutic range (INR above 3.5 or 4.0) after an NSAID under study was added to the acenocoumarol therapy, were compared with patients who did not show such an elevation. A two-sample t-test (average acenocoumarol dosage, age), and chi-square tests (sex, therapeutic range, type of NSAID) were used to test for differences. Genotyping was carried out by analysing blood samples for the relevant CYP2C9 alleles. The study population consisted of 112 patients on stable acenocoumarol therapy, of which 52 (46%) showed an elevation of the INR above the desired therapeutic level (INR 3.5 and 4.0 respectively) after the start of an NSAID under study. In 12 patients, the INR increased above 6. The INR of the other 60 patients (54%) remained constant after the start of one of the NSAIDs under study. There were no statistically significant differences between patients with increased INR and patients without increased INR with regard to age, sex, therapeutic range and average acenocoumarol dosage. Eighty patients, of whom 36 showed an increased INR as a result of a potential acenocoumarol-NSAID drug interaction, were included in the genotyping study. No association between CYP2C9 genotype and an increased INR as a result of the drug-drug interaction was found. In nearly half of a cohort of elderly patients, the INR increased beyond the therapeutic range (INR 3.5 or 4.0) as a result of a potential pharmacokinetic drug-drug interaction between acenocoumarol and diclofenac, naproxen and ibuprofen. The average increase in INR was between 1 and 4. Polymorphism of CYP2C9 does not seem to be a relevant risk factor for the NSAID-acenocoumarol interaction.