This was a 9-month, multicenter, open-label, two-armed, prospective, observational post-authorization survey (PAS), which was conducted by 3,609 general physicians, practitioners and internists (GPs) throughout Germany. This specific study type is regulated by the German Drug Law (AMG) §67(6) and is primarily intended to gather knowledge about the safety and efficacy of marketed drugs in daily practice. The federal panel doctors' associations as well as the higher authorities were duly notified about this investigation. The participating GPs received a small remuneration for the documentation of each patient, which is common practice for this type of PAS. Importantly, as in any PAS, the protocol stipulated no interventions different from routine treatment. In a PAS the drug can only be prescribed in the labeled indication according to the Summary of Product Characteristics (SmPC). The procedures and decisions of the physicians were not influenced and the physicians were completely free to select which patients to treat with the licensed drug being studied, which diagnostic measurements they used, and the way in which they monitored the course of treatment or which concurrent or additional medication they prescribed. Due to the non-interventional type of the study, no ethics committee approval or patient informed consent has to be obtained in accordance with the German local laws and regulations. The participating physicians collected data on the background characteristics of the patients and on key efficacy variables and adverse events (AEs) and documented them in case report forms (CRFs). If any serious adverse events (SAEs) occurred, the GPs were obliged to report them by completing a form within 24 hours and transmitting it to the manufacturer, who then forwarded the report in a standardized format to the relevant authority (Bundesinstitut für Arzneimittel und Medizinprodukte, Bonn, Germany). The collected data, SAE forms and CRFs were not consistently verified in comparison with the source data in the patient files, but the forms were systematically screened for plausibility and completeness.
Patients and Study Conduct
Patients with an indication for treatment with Irbesartan with or without HCTZ were selected by the GPs, using a cohort approach. Only adult patients could be included (18 years), and there were no additional exclusion criteria regarding concomitant medication or concomitant diseases except those specified in the SmPC. Patients were stratified into the metabolic syndrome cohort if they fulfilled the criteria set forth by the National Cholesterol Education Program Expert Panel (NCEP) in 2001 , which applied to 9281 patients. 4919 patients with uncontrolled arterial hypertension without the metabolic syndrome served as controls (the intention was to have a 2 : 1 distribution). The GPs selected patients with uncontrolled arterial hypertension for once daily treatment with Irbesartan (Aprovel™ 75, 150 or 300 mg) as monotherapy or in a combination with 12.5 mg hydrochlorothiazide (HCTZ) (CoAprovel™ 150/12.5 or 300/12.5 Sanofi-Aventis Deutschland GmbH, Berlin). The prescription of additional antihypertensive agents was allowed, as was the discontinuation of other medications, if necessary. There were no stipulations regarding BP targets. However a considerable percentage of German doctors follow the established guidelines of the German Hypertension League, which are < 140/90 mmHg for all patients except patients with diabetes, for whom < 130/80 mmHg applies .
The parameters documented in the study included demographic characteristics (initials, age, sex, weight and height, hip and waist circumference, familial cardiovascular disease and smoking status), medical diagnoses (diabetes, arterial hypertension with the number of years present and known microalbuminuria/proteinuria). Blood pressure was to be taken as a mean of three sphygmomanometric measurements, and pulse rate (beats per minute, bpm) and the respective target blood pressure for each patient were also to be recorded. If available from the charts, the following laboratory parameters were collected: fasting blood glucose, HbA1c, triglycerides, HDL and LDL cholesterol, creatinine and urinary albumin. In addition, concomitant disease was documented: stroke/transient ischemic attack (TIA), neuropathy, coronary heart disease (CHD), heart failure, previous myocardial infarction (MI), aorto-coronary venous bypass operation (ACVB), retinopathy, previous PTCA/stent, left ventricular hypertrophy (LVH), lipid disorders and peripheral arterial disease (PAD). Antihypertensive therapy within the previous 12 months was documented as well as the modifications after switching medications at the baseline visit.
After 3 and 9 months blood pressure measurements were repeated and the following parameters obtained if available: weight, hip and waist circumference, pulse, triglycerides, HDL and LDL cholesterol, fasting blood glucose, HbA1c, creatinine and urinary albumin. Modifications of antihypertensive therapy were documented and whether patients reached blood pressure targets was determined.
The following features of AEs were recorded if these occurred: description, first occurrence, grade of severity, outcome of events (recovered, recovered with sequelae, unresolved), likelihood of causal relationship (possible, probable, improbable, no relationship).
According to the predefined statistical analysis plan, the statistical analysis was performed descriptively and was interpreted in an explorative way. Comparisons were made for blood pressure and proportions of patients with components of the metabolic syndrome positive between baseline and the two post-baseline visits. The absolute and relative frequencies of AEs and the efficacy and tolerability ratings were reported. Post-hoc analyses for subgroups defined by gender, BMI, waist circumference, duration of hypertension, strength of antihypertensive response, and previous and concomitant antihypertensive treatment, respectively, were carried out. The analysis of data was performed with the statistical software package SAS, version 8.2. Test applied are indicated in the legend of tables and figures . Regarding safety, the trial was adequately sized to identify rare AEs, i.e. those that may not have been detected in previous clinical studies, (incidence 1: 1,000) with a probability of > 99% and very rare events (incidence 1: 10,000) with a probability of > 75%.