original article - Nature

ORIGINAL ARTICLE

Management in general practice ofhypertensive patients poorly controlledwith a fixed-dose renin–angiotensinsystem inhibitor and diuretic combination:results from a French national survey

S Ragot1, P Sosner2, C Yau3, P Brunel3 and D Herpin2

1Clinic Research Centre, CHU La Miletrie, Poitiers and University Institute of Public Health, Faculty ofMedicine and Pharmacy, Poitiers, France; 2Department of Cardiology, CHU La Miletrie, Poitiers, France and3Novartis Pharma SAS, Rueil-Malmaison, France

High rates of uncontrolled hypertensives have beenreported in France as well as in other countries, partlyowing to an inadequate management of hypertensionby clinicians. The objectives of the study were (1) todescribe the therapeutic decisions faced by generalpractitioners (GP) in hypertensives not controlled byfixed-dose combination antihypertensive therapy witha renin–angiotensin system (RAS) inhibitor and a diuretic(D) and (2) to evaluate the frequency with whichambulatory blood pressure monitoring (ABPM) is usedin these patients. Each GP had to include five uncon-trolled hypertensives on fixed-dose RAS inhibitor and Dcombination. A total of 2118 GPs included 9551 patients;8643 patients were receiving at least one of the two drugsat a low dose (group 1) and the remaining 908 patients

were receiving high doses of both drugs (group 2).The most common therapeutic choice was that of anew combination, either a fixed-dose in one pill or withseparate preparations: 65% (n¼ 5621) in group 1 and 56%(n¼ 505) in group 2. An increase in dose was chosen in28% of patients in group 1 (n¼ 2467) and continuation oftreatment without modification in 27% of patientsin group 2 (n¼ 242). A third active principle was onlyadded in 2.5% of patients in group 1 (n¼ 219) and 11% ingroup 2 (n¼ 103). Ambulatory blood pressure monitoringwas employed in 25% of patients (n¼ 2413). An improve-ment in hypertension management should be expectedfrom the implementation of the 2005 French guidelines.Journal of Human Hypertension (2006) 20, 407–418.doi:10.1038/sj.jhh.1002000; published online 30 March 2006

Keywords: renin–angiotensin system inhibitor; uncontrolled hypertension; ambulatory BP measurement; therapeuticstrategies for hypertension; general practice; fixed-dose combination antihypertensive therapy

Introduction

Although cardiovascular morbidity and mortalityare lower in France than in Anglo-Saxon andEastern countries, arterial hypertension remainsa major Public Health problem, largely owing to itshigh incidence. A recent survey1 estimated thateight million French people are undergoing treat-ment for hypertension (equivalent to 24% of theadult population). Moreover, around 80% of hyper-tensives are thought to be treated.2 A high propor-tion of hypertensives are poorly controlled despite

their treatment; Chamontin et al.3 noted a particu-larly high failure rate (65–75%) in their study.4–6

These poor results could be partly explained byinadequate management of antihypertensive treat-ments by clinicians. The French study, PHENOM-EN,4 highlighted the discrepancy between thevigour of the antihypertensive treatment and theobserved blood pressure (BP) control: 45% ofpatients with uncontrolled hypertension undertreatment were receiving mono- or dual antihyper-tensive therapy. Similarly, the PRATIK study5

demonstrated the under-use of combination thera-pies (in less than 2/3 high-risk hypertensives) andthe suboptimal utilization of triple therapies (whichin 25% of cases did not include diuretics).

French guidelines for the treatment of hyperten-sive patients, employed at the time of this study,6

propose first intention antihypertensive treatmentwith single drug or fixed-dose combination therapy

Received 3 December 2005; revised 10 January 2006; accepted 12January 2006; published online 30 March 2006

Correspondence: Dr S Ragot, Institut Universitaire de SantePublique, Faculte de Medecine et de Pharmacie, 34 Rue du Jardindes Plantes, Poitiers 86 000, France.E-mail: [email protected]

Journal of Human Hypertension (2006) 20, 407–418& 2006 Nature Publishing Group All rights reserved 0950-9240/06 $30.00

www.nature.com/jhh

along with lifestyle and dietary measures: efficacyand tolerability to be evaluated over the first 2months of treatment. When the first drug is welltolerated but has an insufficient antihypertensiveeffect, the preferred course of action should be theaddition of a second active principle, preferably athiazide if the first active principle does not belongto this class. These guidelines are less helpful whenit comes to treating cases that are resistant to dualtherapy.

Three strategic choices are available to theclinician: change the dual therapy; add a thirdactive principle; or increase the dose of one or bothcomponents of the combination if the dose was notoptimal.

To our knowledge, there are no published data onthe attitude of general practitioners (GP) in Francewhen confronted with this therapeutic decision.The present survey was thus designed to find outmore about the decisions of French GP in the eventof treatment failure with a dual therapy combining,in a fixed-dose form, a renin–angiotensin system(RAS) inhibitor: either an angiotensin-convertingenzyme inhibitor (ACEI) or angiotensin II receptorblocker (ARB) combined with a diuretic. This is themost widely prescribed combination of antihyper-tensive drugs in France (47% of patients); in halfof the patients, it is prescribed as a fixed-dosecombined pill.7

Angiotensin-converting enzyme inhibitors wereintroduced in France in the 1980s and the first ARBin 1995. Their efficacy in terms of reducing BP iscomparable to other classes of antihypertensive, butthe ARB is thought to be better tolerated. Themorbidity/mortality benefits of the two classes arenow well documented in different populations ofhypertensives.8–12

In the event of clinically judged resistance to adual therapy, practitioners may also elect to sub-stantiate the impression by using a method of BPmeasurement that eliminates the ‘white-coat’ effect,such as self-measurement or ambulatory BP mea-surement (ABPM). We also lack data on thefrequency of use of these methods in such circum-stances.

The primary objective of this survey was toexplore the therapeutic decisions made by GP forhypertensive patients who are poorly controlled byfixed-dose antihypertensive therapy combining aRAS inhibitor and a diuretic.

The secondary objectives were to determine theclinical characteristics that influence the therapeu-tic decisions and to evaluate the frequency of use ofABPM in uncontrolled hypertensives.

Materials and methods

This cross-sectional observational study was con-ducted on a representative population of 2500French GP selected at random from an independent

database. Each doctor participating in the study hada 6-month period after the start date to include thefirst five patients seen in consultation who fulfilledthe inclusion criteria: patients over 18 years of agewith uncontrolled essential hypertension (BPX140/90mm Hg) despite treatment for at least 6 months, atthe same dose, with a fixed-dose therapy combininga RAS inhibitor (ACEI or ARB) and a thiazidediuretic. Patients with free combination therapy (i.e.the two active principles not combined in the samepill) and those receiving a fixed-dose combinationtherapy specifically indicated for the first intentiontreatment of hypertension were not included.

The patients were informed about the study and itwas made clear that their participation in the studywould not affect their treatment or follow-up; thedoctors were free to choose their prescriptions andmethods of follow-up.

Data collectionDuring the consultation, BP and heart rate weremeasured in the sitting position after a 5-min restperiod. Three measurements were taken at 1-minintervals and only the last two were taken intoaccount.13 The class and dose of the two drugs in thecurrent antihypertensive treatment were recorded,as well as the date on which this treatment wasinitiated.

The doctor had to record (three options: yes, no,don’t know) the following associated cardiovascularpathologies and risk factors: active smoking, dia-betes, dyslipidaemia (total cholesterol 42 g/l and/ortriglycerides 42 g/l), history of coronary disease,history of a stroke or transient ischaemic accident,history of lower limb arteriopathy, renal insuffi-ciency (serum creatinine 420mg/l or 4177 mmol/l)and proteinuria.

They also had to indicate whether an ABPM hadbeen performed during the last 12 months or wasplanned in the near future.

Finally, for each patient, the doctor had to recordwhich of the following therapeutic decisions hewould take: no change in treatment, change of dose(of diuretic, of RAS inhibitor or of both drugs),addition of a third active principle, discontinuationof current dual therapy and prescription of a newfixed-dose or free combination. If the doctor decidednot to change the treatment, he had to indicatewhether this choice was influenced by the toler-ability of the treatment and/or refusal on the part ofthe patient to change, and/or an attempt to limit thenumber of doses of medication.

Data analysisStatistical analysis was performed using SASs

software version 8.2 (Institute Inc., Cary, NC,USA). Questionnaires with data missing for BP onconsultation, the nature and/or dose of treatments atinclusion, and the therapeutic decision were not

Management of uncontrolled hypertensionS Ragot et al

408

Journal of Human Hypertension

included in the analysis. Similarly, patients pre-senting with a BP (mean of the second and thirdmeasurements taken during the consultation) ofo140/90mm Hg were excluded from the analysis.

The quantitative data were described using meansand standard deviation, and the categorical data byan overall number and the corresponding percen-tage. The description and analysis of the therapeuticdecisions was classified as a function of currentdual therapy and divided into two subgroups: group1¼patients on dual therapy at submaximal doses(including group 1A¼ low-dose RAS inhibi-torþ low-dose diuretic, group 1B¼high-dose RASinhibitorþ low-dose diuretic, group 1C¼ low-doseRAS inhibitorþhigh-dose diuretic) and group2¼patients with dual therapy at maximal doses.

The therapeutic decision was summarized as a‘non-conservative approach’ (prescription of a newcombined or separate drug combination) or ‘con-servative approach’ (no modification of treatmentor change in dose or addition of a third activeprinciple).

To investigate factors linked to the therapeuticdecision or the use of an ABPM, quantitativevariables were compared between the groups byanalysis of variance (42 groups) or Student’s t-test(two groups). The w2 test was used for the categoricalvariables.

Multivariate analyses (logistic regression) wereperformed to identify factors that could indepen-dently explain the therapeutic decision or the useof ABPM. All of the variables with a P-value o0.15were retained, and a stepwise selection procedureusing entrance and exit thresholds of 0.05 wasapplied. A P-value o0.05 was considered signifi-cant.

Justification for the number of patientsThe main objective of this survey was to define, withsufficient precision, the therapeutic decision as afunction of the treatment of patients at consultation.With a population of 8500 patients in the grouptaking low doses of RAS inhibitor or diuretics, theprecision was 1.1% for a treatment modality with afrequency of 50 and 0.6% if the frequency was 10%.With a population of 1000 patients in the grouptaking high doses of RAS inhibitor and diuretics, theprecision was 3.1% for a treatment modality with afrequency of 50 and 1.9% if the frequency was 10%.Assuming a fall-out of 20% (dossiers not returned orunusable), a total of 12 500 patients were required.

Results

Of the 2500 selected GP, 2118 doctors participatedin the study (mean age 4977 years; 30–73 years) andrecruited 11 118 patients. Of these 11 118 patients,254 patients presented with a mean BP of o140/90mm Hg, and in 1313 patients, at least one of the

following variables was missing: BP values atconsultation (17 missing), nature and/or dose oftreatments on inclusion (1296 missing), including25 patients whose therapeutic decision was alsomissing.

Thus, 9551 patients (85.9%) were included in theanalysis. Mean age was 64711 years. There were5069 men (53.1%) and 4475 women (46.9%); genderwas not indicated for seven patients.

The most common current antihypertensive treat-ment was the combination of low-dose RAS inhi-bitor with low-dose diuretic (5753 patients or 60.2%of the analysed population; group 1A). Next camethe combination of high-dose RAS inhibitor andlow-dose diuretic (2694 patients or 28.2% of theanalysed population; group 1B), followed by thecombination of high doses of both types of drug(9.5%, n¼ 908, group 2). The combination of low-dose RAS inhibitor with high-dose diuretic was onlynoted in 2.1% of patients (n¼ 196, group 1C).

The RAS inhibitor used was ARB in 54.6% ofcases (n¼ 5212).

The demographic and clinical characteristics ofthe whole population and of groups 1 and 2 arelisted in Table 1. The distribution of BPs accordingto their ANAES grades demonstrated a higherproportion of grade 1 and grade 3 hypertensives ingroup 2 than in subgroups 1A, 1B or 1C (Figure 1).

The therapeutic decisions in the four treatmentgroups are illustrated in Figure 2. Confronted withtreatment failure, the practitioners’ preferred ap-proach was the prescription of a new combinedfixed-dose or separate combination, regardless of thetype of previous treatment: in 65.0% (n¼ 5621) ofpatients who were not receiving maximal dosesof dual therapy (group 1) and in 55.6% (n¼ 505) ofhypertensives receiving maximal dual therapy(group 2). Modification of the dose was chosen in28.5% (n¼ 2467) of patients in group 1 whereas athird active principle was rarely prescribed (2.5%;n¼ 219).

In group 2, the continuation of treatment withoutmodification was the second commonest strategy(26.7% of patients) whereas a third active principlewas only added in 11.3% of patients. A change indose was indicated in 6.4% of patients, equivalentto 58 patients, 31 of whom presented with a poortolerability.

Thus, the ‘non-conservative’ approach, that is, thechoice of a new combined or separate drug combi-nation, was more common than the conservativeapproach, notably in groups 1B (77.0% of cases,n¼ 2075) and 1C (70.4% of cases, n¼ 138), and in asmaller proportion of groups 1A (59.2%, n¼ 3408)and 2 (55.6%, n¼ 505).

The effects of the patients’ demographic andclinical characteristics (BP, cardiovascular risk fac-tors and concurrent clinical pathologies) and theclass of the RAS inhibitor used on the doctor’stherapeutic decision are shown in Table 2a and b forgroups 1 and 2, respectively.


409


The factors influencing a non-conservative ap-proach (vs conservative) retained in the multivariateanalysis are listed in Table 3a and b. Thus, in group1, the non-conservative approach was indepen-

dently related to the highest BPs, to systolic–diastolic (SD) hypertension, to the use of an ACEIrather than an ARB and to the duration and dosageof the current dual therapy.

Table 1 Demographic and clinical characteristics as a function of the dose of the RAS inhibitor – diuretic combination

Group 1Low-dose RAS inhibitor or

diuretic (n¼8643)

Group 2High-dose RAS inhibitor and

diuretic (n¼908)

Total(n¼ 9551)

Age (years) 64.1711.1 64.6711.3 64.1711.1SexMale 4537 (52.5%) 532 (58.7%) 5069 (53.1%)Female 4100 (47.5%) 375 (41.3%) 4475 (46.9%)

Body mass index (kg/m2)o25 2443 (28.4%) 231 (25.6%) 2674 (28.1%)Between 25 and 30 4140 (48.1%) 401 (44.4%) 4541 (47.7%)430 2029 (23.6%) 272 (30.1%) 2301 (24.2%)

Duration of hypertension treatmentLess than 3 months 1668 (19.8%) 252 (29.2%) 1920 (20.7%)More than 3 months 6744 (80.2%) 611 (70.8%) 7355 (79.3%)

Blood pressure (mm Hg)Systolic 160.3710.7 158.5712.6 160.1710.9Diastolic 92.877.3 90.778.7 92.677.5

Isolated systolic hypertension 1755 (20.3%) 301 (33.1%) 2056 (21.5%)

Active smokingYes 2013 (23.4%) 208 (23.0%) 2221 (23.3%)No 6569 (76.2%) 688 (76.1%) 7257 (76.2%)Don’t know 50 (0.6%) 8 (0.9%) 58 (0.6%)

DiabetesYes 1916 (22.2%) 228 (25.2%) 2144 (22.5%)No 6682 (77.5%) 675 (74.7%) 7357 (77.3%)Don’t know 30 (0.3%) 1 (0.1%) 31 (0.3%)

DyslipidaemiaYes 5390 (62.5%) 569 (62.9%) 5959 (62.6%)No 3176 (36.8%) 332 (36.7%) 3508 (36.8%)Don’t know 61 (0.7%) 4 (0.4%) 65 (0.7%)

History of coronary diseaseYes 1093 (12.7%) 128 (14.2%) 1221 (12.8%)No 7424 (86.1%) 771 (85.3%) 8195 (86.1%)Don’t know 104 (1.2%) 3 (0.3%) 107 (1.1%)

History of stroke or TIAYes 594 (6.9%) 74 (8.2%) 668 (7.0%)No 7932 (92.0%) 819 (90.6%) 8751 (91.9%)Don’t know 99 (1.1%) 9 (1.0%) 108 (1.1%)

History of lower limb arteriopathyYes 521 (6.0%) 67 (7.4%) 588 (6.2%)No 8011 (92.9%) 834 (92.3%) 8845 (92.9%)Don’t know 93 (1.1%) 5 (0.6%) 98 (1.0%)

Renal insufficiencyYes 148 (1.7%) 28 (3.1%) 176 (1.8%)No 8381 (97.2%) 870 (96.2%) 9251 (97.1%)Don’t know 94 (1.1%) 6 (0.7%) 100 (1.1%)

ProteinuriaYes 295 (3.4%) 37 (4.1%) 332 (3.5%)No 7640 (88.6%) 790 (87.4%) 8430 (88.5%)Don’t know 684 (7.9%) 77 (8.5%) 761 (8.0%)

Abbreviations: RAS, renin–angiotensin system; TIA, transient ischaemic accident.


410


In the group of patients receiving the high-dosecombination (group 2), multivariate analysis demon-strated that the severity of the hypertension (grade2 vs grade 1), the SD hypertension type, the useof an ACEI and a duration of fixed-dose combinationtherapy of more than 3 months were factorsinducing a non-conservative approach.

The doctors declared that ABPM had been con-ducted in the last 12 months or was planned in thenear future in 2413 patients, equivalent to 25.3% ofthe study population. The proportion of patients

undergoing ABPM was similar in all treatmentgroups: 26.0% in group 1A (n¼ 1495), 23.0% ingroup 1B (n¼ 619), 33.2% in group 1C (n¼ 65) and25.8% in group 2 (n¼ 234). As shown in Table 4,after adjustment, ABPM was requested more often inyounger patients, those with higher BP figures, thoseundergoing fixed-dose combination therapy formore than 3 months, cases with SD hypertensionand in the presence of associated risk factors orclinical pathologies (dyslipidaemia, diabetes, smok-ing, history of stroke, transient ischaemic attack orcoronary disease).

Finally, in the subgroup of patients whose treat-ment was not altered, the doctors cited the followingfactors as influencing this therapeutic decision:group 1: tolerability in 192 cases (57.1%), patient’srefusal to change in 78 cases (23.2%) and a desire tolimit the number of doses in 85 cases (25.3%). Ingroup 2, the distribution was as follows: tolerabilityfor 186 cases (76.9%), patient’s refusal to change in23 cases (9.5%) and a desire to limit the number ofdoses in 88 cases (36.4%).

Of note, 285 patients (3.0%) spontaneously re-ported at least one adverse event during the visit:241 patients in group 1 (2.8%) and 44 patients ingroup 2 (4.8%).

Discussion

Uncontrolled hypertensionThe results of the treatment of hypertensive patientsin France are relatively disappointing.3,5,7 Thus, outof the hypertensives receiving treatment in theFrench subgroup of the MONICA study,14 whichincluded all types of patients (35–64 years of age),the 140/90mm Hg goal was only attained in men in21% of the 35 to 44-year-old range, 11% of the 45–54age group and 8% of the 55–64 age group. In women,the control rate was 28, 16 and 15% in these threeage groups, respectively. The IHPAF survey15 ofpatients, followed up by the French medical servicefor health and safety in the workplace, confirmedthis poor rate of BP control in treated hypertensives:only 13% of men and 33% of women attained theobjective. A rate of 24% was reported in a surveycarried out in general practice.3

A similar failure rate has been reported in othercountries: the rate of control is less than 6% in theUnited Kingdom,16 50% in Scotland but with atarget of 160/95mm Hg17 and 31% in the UnitedStates. In the latter, poor results were noted moreoften in women, Mexican-Americans and patientsover 60 years of age.18

There are several reasons for these poor results:19

� Some are linked to the patients themselves andinclude a lack of motivation, poor adherence, pooradaptation to secondary effects, even very mildones, excess salt consumption or excessive alco-hol intake, etc.

Figure 1 Distribution of blood pressure values categorizedaccording to the ANAES 2000 grades, as a function of the doseof current dual therapy. Groups: group 1A¼ low-dose renin–angiotensin system (RAS) inhibitorþ low-dose diuretic; group1B¼high-dose RAS inhibitorþ low-dose diuretic; group1C¼ low-dose RAS inhibitorþhigh-dose diuretic; group2¼high-dose RAS inhibitorþhigh-dose diuretic. ANAES 2000grades: & Grade 1, grade 2, ’ grade 3. Figures inside the barsindicate absolute number of patients. Between-group comparison:Po0.001.

Figure 2 Therapeutic decisions of the doctor, as a function of thedose of dual therapy. Top left: group 1A¼ low-dose renin–angiotensin system (RAS) inhibitorþ low-dose diuretic; bottomleft: group 1B¼high-dose RAS inhibitorþ low-dose diuretic; topright: group 1C¼ low-dose RAS inhibitorþhigh-dose diuretic;bottom right: group 2¼high-dose RAS inhibitorþhigh-dosediuretic. Circles: therapeutic decisions, & no change, changeof dose, ’ addition of a third active principle, newcombination. Bars: type of drug dose change. blocker,diuretic, blocker and diuretic. Figures indicate absolute numberof patients.


411


Table 2 Therapeutic decision of the doctor as a function of the demographic and clinical characteristics of the patients and according tothe class of RAS inhibitor used in groups 1 and 2: univariate approach

Conservative approach

Continuation oftreatment without

modification

Addition of athird activeprinciple

Change ofdose

Sum of the threedecisions

Non-conservativeapproach

P, conservativevs non-conser-vative approach

(a) Group 1SexMale 167 (3.7%) 117 (2.6%) 1333 (29.4%) 1617 (35.6%) 2920 (64.4%) NSFemale 169 (4.1%) 102 (2.5%) 1132 (27.6%) 1403 (34.2%) 2697 (65.8%)

Age (years)Less than 55 63 (3.5%) 49 (2.7%) 552 (30.8%) 664 (37.0%) 1131 (63.0%) NS55–65 93 (3.8%) 59 (2.4%) 715 (28.9%) 867 (35.1%) 1603 (64.9%)Over 65 180 (4.1%) 111 (2.5%) 1200 (27.4%) 1491 (34.1%) 2887 (65.9%)

Body mass index (kg/m2)o25 134 (5.5%) 65 (2.7%) 704 (28.8%) 903 (37.0%) 1540 (63.0%) NSBetween 25 and 30 143 (3.5%) 88 (2.1%) 1181 (28.5%) 1412 (34.1%) 2728 (65.9%)430 58 (2.9%) 65 (3.2%) 580 (28.6%) 703 (34.6%) 1326 (65.4%)

Duration of dual therapyLess than 3 months 84 (5.0%) 49 (2.9%) 599 (35.9%) 732 (43.9%) 936 (56.1%) o0.001More than 3 months 239 (3.5%) 166 (2.5%) 1793 (26.6%) 2198 (32.6%) 4546 (67.4%)

Severity of BP valuesGrade 1 246 (7.6%) 79 (2.4%) 998 (30.9%) 1323 (41.0%) 1905 (59.0%) o0.001Grade 2 81 (1.7%) 121 (2.5%) 1308 (27.4%) 1510 (31.6%) 3264 (68.4%)Grade 3 9 (1.4%) 19 (3.0%) 161 (25.1%) 189 (29.5%) 452 (70.5%)

Type of hypertensionIsolated systolic 136 (7.7%) 57 (3.2%) 536 (30.5%) 729 (41.5%) 1026 (58.5%) o0.001Systolic–diastolic 200 (2.9%) 162 (2.4%) 1931 (28.0%) 2293 (33.3%) 4595 (66.7%)

SmokingYes 66 (3.3%) 54 (2.7%) 597 (29.7%) 717 (35.6%) 1296 (64.4%) NSNo 264 (4.0%) 165 (2.5%) 1858 (28.3%) 2287 (34.8%) 4282 (65.2%)

DiabetesYes 51 (2.7%) 66 (3.4%) 548 (28.6%) 665 (34.7%) 1251 (65.3%) NSNo 280 (4.2%) 152 (2.3%) 1909 (28.6%) 2341 (35.0%) 4341 (65.0%)

DyslipidaemiaYes 164 (3.0%) 143 (2.7%) 1527 (28.3%) 1834 (34.0%) 3556 (66.0%) o0.05No 164 (5.2%) 75 (2.4%) 912 (28.7%) 1151 (36.2%) 2025 (63.8%)

History of coronary diseaseYes 28 (2.6%) 31 (2.8%) 320 (29.3%) 379 (34.7%) 714 (65.3%) NSNo 300 (4.0%) 186 (2.5%) 2103 (28.3%) 2589 (34.9%) 4835 (65.1%)

History of stroke or TIAYes 16 (2.7%) 23 (3.9%) 155 (26.1%) 194 (32.7%) 400 (67.3%) NSNo 312 (3.9%) 195 (2.5%) 2271 (28.6%) 2778 (35.0%) 5154 (65.0%)

History of lower limb arteriopathyYes 15 (2.9%) 18 (3.5%) 165 (31.7%) 198 (38.0%) 323 (62.0%) NSNo 313 (3.9%) 198 (2.5%) 2263 (28.2%) 2774 (34.6%) 5237 (65.4%)

Renal insufficiencyYes 8 (5.4%) 5 (3.4%) 44 (29.7%) 57 (38.5%) 91 (61.5%) NSNo 321 (3.8%) 213 (2.5%) 2386 (28.5%) 2920 (34.8%) 5461 (65.2%)

ProteinuriaYes 10 (3.4%) 10 (3.4%) 82 (27.8%) 102 (34.6%) 193 (65.4%) NSNo 290 (3.8%) 190 (2.5%) 2198 (28.8%) 2678 (35.1%) 4962 (64.9%)

Type of RAS inhibitorARB 212 (4.6%) 124 (2.7%) 1824 (39.6%) 2160 (46.9%) 2441 (53.1%) o0.001ACEI 124 (3.1%) 95 (2.4%) 643 (15.9%) 862 (21.3%) 3179 (78.7%)


412


(b) Group 2SexMale 138 (25.9%) 62 (11.7%) 29 (5.5%) 229 (43.0%) 303 (57.0%) NSFemale 104 (27.7%) 41 (10.9%) 28 (7.5%) 173 (46.1%) 202 (53.9%)

Age (years)Less than 55 50 (27.8%) 23 (12.8%) 11 (6.1%) 84 (46.7%) 96 (53.3%) NS55–65 50 (20.4%) 27 (11.0%) 17 (6.9%) 94 (38.4%) 151 (61.6%)Over 65 142 (29.4%) 53 (11.0%) 30 (6.2%) 225 (46.6%) 258 (53.4%)

Body mass index (kg/m2)o25 60 (26.0%) 27 (11.7%) 21 (9.1%) 108 (46.8%) 123 (53.2%) NSBetween 25 and 30 120 (29.9%) 37 (9.2%) 26 (6.5%) 183 (45.6%) 218 (54.4%)430 60 (22.1%) 39 (14.3%) 11 (4.0%) 110 (40.4%) 162 (59.6%)

Duration of dual therapyLess than 3 months 132 (52.4%) 34 (13.5%) 14 (5.6%) 180 (71.4%) 72 (28.6%) o0.001More than 3 months 94 (15.4%) 63 (10.3%) 36 (5.9%) 193 (31.6%) 418 (68.4%)

Severity of BP valuesGrade 1 198 (46.3%) 33 (7.7%) 21 (4.9%) 252 (58.9%) 176 (41.1%) o0.001Grade 2 41 (10.3%) 51 (12.8%) 31 (7.8%) 123 (31.0%) 274 (69.0%)Grade 3 3 (3.6%) 19 (22.9%) 6 (7.2%) 28 (33.7%) 55 (66.3%)

Type of hypertensionIsolated systolic 158 (52.5%) 16 (5.3%) 14 (4.7%) 188 (62.5%) 113 (37.5%) o0.001Systolic–diastolic 84 (13.8%) 87 (14.3%) 44 (7.2%) 215 (35.4%) 392 (64.6%)

SmokingYes 53 (25.5%) 23 (11.1%) 10 (4.8%) 86 (41.3%) 122 (58.7%) NSNo 184 (26.7%) 80 (11.6%) 48 (7.0%) 312 (45.3%) 376 (54.7%)

DiabetesYes 52 (22.8%) 33 (14.5%) 13 (5.7%) 98 (43.0%) 130 (57.0%) NSNo 189 (28.0%) 69 (10.2%) 45 (6.7%) 303 (44.9%) 372 (55.1%)

DyslipidaemiaYes 144 (25.3%) 67 (11.8%) 32 (5.6%) 243 (42.7%) 326 (57.3%) NSNo 97 (29.2%) 35 (10.5%) 24 (7.2%) 156 (47.0%) 176 (53.0%)

History of coronary diseaseYes 35 (27.3%) 15 (11.7%) 10 (7.8%) 60 (46.9%) 68 (53.1%) NSNo 203 (26.3%) 88 (11.4%) 47 (6.1%) 338 (43.8%) 433 (56.2%)

History of stroke or TIAYes 16 (21.6%) 12 (16.2%) 5 (6.8%) 33 (44.6%) 41 (55.4%) NSNo 222 (27.1%) 91 (11.1%) 51 (6.2%) 364 (44.4%) 455 (55.6%)

History of lower limb arteriopathyYes 19 (28.4%) 8 (11.9%) 2 (3.0%) 29 (43.3%) 38 (56.7%) NSNo 220 (26.4%) 95 (11.4%) 55 (6.6%) 370 (44.4%) 464 (55.6%)

Renal insufficiencyYes 4 (14.3%) 3 (10.7%) 3 (10.7%) 10 (35.7%) 18 (64.3%) NSNo 236 (27.1%) 99 (11.4%) 55 (6.3%) 390 (44.8%) 480 (55.2%)

ProteinuriaYes 6 (16.2%) 8 (21.6%) 1 (2.7%) 15 (40.5%) 22 (59.5%) NSNo 217 (27.5%) 82 (10.4%) 52 (6.6%) 351 (44.4%) 439 (55.6%)

Nature of the RAS inhibitorARB 227 (32.7%) 95 (15.5%) 42 (6.9%) 364 (59.6%) 247 (40.4%) o0.001ACEI 15 (5.1%) 8 (2.7%) 16 (5.4%) 39 (13.1%) 258 (86.9%)

Abbreviations: ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; RAS, renin–angiotensin system; TIA,transient ischaemic accident.

Table 2 Continued

Conservative approach

Continuation oftreatment without

modification

Addition of athird activeprinciple

Change ofdose

Sum of the threedecisions

Non-conservativeapproach

P, conservativevs non-conser-vative approach


413


� Less commonly, failure may be caused by thespecificity of the arterial hypertension: this is thecase with low-renin hypertension or renal insuffi-ciency associated with hypertension.

� Doctors may also be more directly responsiblefor poor BP control: measurement of clinic BPmay not be accurate and may need to besupplemented by ABPM or BP self-measurement.They may also have prescribed antagonistic drugs,particularly non-steroidal anti-inflammatoryagents. Finally, they may have used suboptimaltherapeutic regimens (doses that are too low,

inconsistent combinations, under-use of tripletherapies, etc.).

This last point was examined in particular depthin our study, the primary aim of which was toanalyse the customary attitudes of practitioners,which is an essential starting point for the instiga-tion of properly adapted training programmes.

Furthermore, this study provides us with adescription of the profile of a sample of non-normalized patients; of particular note is the markedprevalence in our survey of the combination of

Table 3 Factors involved in the therapeutic decision of the doctor in groups 1 and 2: results of the multivariate analysis

Dependent variable¼ two classes of therapeutic approach: non-conservative vs conservative OR CI of the OR P-value

(a) Group 1 (n¼ 8412 patients)Treatment before consultation(reference class: low-dose RAS inhibitor/low-dose diuretic)High-dose RAS inhibitor+low-dose diuretic 1.380 [0.994; 1.915] 0.06Low-dose RAS inhibitor+high-dose diuretic 2.347 [2.102; 2.620] o0.0001

Grade of hypertension (reference class: grade 1)Grade 2 1.437 [1.300; 1.588] o0.0001Grade 3 1.637 [1.343; 1.994] o0.0001

Isolated systolic hypertension (reference class: SD hypertension) 0.768 [0.683; 0.863] o0.0001Nature of the RAS inhibitor: ARB (reference class: ACEI) 0.314 [0.284; 0.346] o0.0001Duration of treatment before consultation: 43 months (reference class: p3 months) 1.489 [1.326; 1.673] o0.0001

(b) Group 2 (n¼ 858 patients)Grade of hypertension (reference class: grade 1)Grade 2 2.030 [1.426; 2.888] 0.03Grade 3 1.789 [0.985; 3.138] NS

Isolated systolic hypertension (reference class: systolic–diastolic HT) 0.483 [0.337; 0.692] o0.0001Nature of the RAS inhibitor: ARB (reference class: ACEI) 0.137 [0.091; 0.204] o0.0001Duration of treatment before consultation: 43 months (reference class: p3 months) 3.616 [2.525; 5.180] o0.0001

Abbreviations: ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; HT, hypertension; RAS, renin–angiotensinsystem; SD hypertension, systolic–diastolic hypertension.Only the significant factors were included in this table; the following factors were not significant: sex, age, BMI, smoking habit, diabetes,dyslipidaemia, history of coronary disease, stroke, TIA, lower limb arteriopathy, renal insufficiency and proteinuria.

Table 4 Factors correlated with the use of ABPM in the study population

N¼9275 OR CI of the OR P-value

Age (years) (reference class: o55) o0.000155–65 0.751 [0.659; 0.856]Over 65 0.554 [0.488; 0.629]

Grade of hypertension (reference class: grade 1) o0.0001Grade 2 1.143 [1.031; 1.266]Grade 3 1.619 [1.353; 1.938]

Duration of treatment before consultation 43 months (vs p3 months) 0.696 [0.622; 0.778] o0.0001Isolated systolic hypertension (vs SD hypertension) 0.803 [0.711; 0.907] 0.0004Smoking 1.157 [1.033; 1.296] 0.0115Diabetes 1.149 [1.026; 1.287] 0.0165Dyslipidaemia 1.376 [1.241; 1.526] o.0001History of coronary disease 1.290 [1.121; 1.484] 0.0004History of stroke or TIA 1.379 [1.151; 1.653] 0.0005

Abbreviations: ABPM, ambulatory blood pressure monitoring; SD, systolic–diastolic; TIA, transient ischaemic accident.Only the significant factors were included in this table; the following factors were not significant: sex, BMI, renal insufficiency, proteinuria,arteriopathy and treatment. A total of 9275 patients were included in the model (i.e. no missing variables).


414


hypertension and dyslipidaemia (63% of patients),whereas 28% of patients in the FLAHS study1 weretreated for both hypertension and dyslipidaemia.Similarly, diabetes was associated with hyperten-sion in 22% of cases in our population, comparedwith only 8% of cases in the FLAHS survey. Wetherefore observed, in common with other authors,a higher prevalence of associated risk factors inuncontrolled hypertensives. In the experience ofAsmar et al.,20 based on 61108 patients seen inhealthcare centres, elevated BP was associated withat least one other risk factor (most commonlyhypercholesterolemia) in 84% of men and 77% ofwomen, and with at least two other risk factors in23% of men and 10% of women. Amar et al.5

demonstrated that the control of hypertensiondecreased as the number of risk factors increased:only 27% of patients were controlled in the grouppresenting with at least three associated risk factorsor target organ damage, compared to 43% in thegroup without associated risk factors.

Management of resistance to combined therapyIn our survey, doctors were asked what they mightdo with a particular patient they had just seen.Clearly, faced with a failure of combined therapy,they favoured the use of a new fixed-dose orseparate combination, that is, a therapeutic ‘switch’.The latter was chosen in 65% of patients who werenot receiving maximal doses of the dual therapy,and in 56% of those who were receiving maximaldoses of both drugs.

This therapeutic decision, which can be describedas ‘non-conservative’, was more common in thepatients with the highest BP, those with SDhypertension, those whose dual therapy includedan ACEI and those whose treatment was long-standing. We also found that in patients onsubmaximal doses of combined therapy, the actualdosages had an influence, with a new combinationbeing chosen more often if one of the drugs of thedual therapy was at its maximal dose. Associatedrisk factors and clinical pathologies were not foundto influence the doctors’ decision.

Among the various forms of conservative ap-proach, dosage was altered in 28% of patientsreceiving submaximal doses, whereas a third activeprinciple was rarely added in this subgroup (2.5%).In patients receiving maximal doses, a third activeprinciple was also little prescribed (11% in thisgroup).

Arguments for and against each of the variousstrategies can be found in the literature.

Increasing the doses. This is not always the bestsolution, as it is often associated with an exacerba-tion of secondary effects, and not necessarily withany substantial increase in antihypertensive action.

Law et al.,21 pooling the results of 354 randomizedtrials, demonstrated the heterogenicity of the dose–

tolerance relationship between the different classes:although an increase in dosage is strongly correlatedwith an increase in secondary effects for thiazidediuretics, beta-blockers and calcium channel inhi-bitors, it does not alter the secondary effects of ACEI(principally coughing), whereas ARB are equallywell tolerated at high and low doses.

In a population of hypertensive patients over 65years of age, Morgan et al.22 demonstrated that incomparison with a single dose, doubling the dosewas only accompanied by a significant additionaldecrease in BP with amlodipine (5mg), but not withatenolol (25mg), enalapril (20mg) or hydrochloro-thiazide (HCTZ, 25mg). It should be noted thatthe 25mg dose of HCTZ, considered as single byMorgan, is in fact the highest dose available in fixed-dose combinations. Other authors have demon-strated that this 25mg dose was significantly moreeffective than the 12.5mg dose at lowering BP, andalso note the absence of any substantial increaseabove 50mg.23 It should also be noted that it is onlyabove 25mg that HCTZ significantly increases basalplasma renin activity.24 In a previous study, wedemonstrated that the prescription of a combinationof valsartanþ 25mg HCTZ in patients who were notcontrolled by a combination of ARBþ 12.5mg HCTZnormalized an additional 23% of patients.25 In thepresent study, the majority of doctors who chose toincrease the dose in patients receiving the two drugsat low doses increased the dose of both the RASinhibitor and the diuretic.

Treatment switch. The advantage of switchingtreatments after a monotherapy has been welldocumented by Dickerson et al.,26 who demon-strated in a crossover study that patients whoresponded to diuretics and calcium channel block-ers were little or not at all responsive to ACEI andbeta-blockers, and vice versa. This concept has nowbeen incorporated into the BHS guidelines.27 How-ever, switching in the event of dual therapy failureis only mentioned in the literature in combinationwith a change in the dose of one of the components,as indicated above.25

It is difficult to identify practitioners’ motives forswitching treatment. Thus, in a population-basedstudy on the renewal of prescriptions at 12 monthsin newly treated hypertensives, Bloom28 recorded acontinuation of treatment in 64% of cases with ARB,58% with ACEI, 50% with calcium channel block-ers, 43% with beta-blockers and 38% with thethiazides. However, these figures had no correlationwith the switch rate, which was between 6 and 9%for each of these drug classes. A recent Canadianstudy gave similar results for continuation oftreatment at 90 days.29 However, in a large Taiwa-nese study, with more than 500 000 patients, theswitch rate was found to be much greater than inBloom et al.’s experience: 29% with ACEI anddiuretics, 23% with beta-blockers and 21% for thecalcium channel blockers.30 A switch was therefore


415


the most common therapeutic decision of thedoctors in our survey, given that a switch wasdefined as a change to another therapeutic class or achange of drug within the same class. The link wefound between ‘non-conservative approach’ and SDhypertension could illustrate that patients with SDhypertension are considered by GP at higher riskthan those with isolated systolic hypertension. Sucha misconception has been frequently reported in theliterature.31

The addition of a third active principle. Lawet al.21 have demonstrated the advantages of thisstrategic choice. They found that a synergistic tripletherapy at half-doses was not only significantlymore effective on BP than a dual therapy prescribedat half-doses (�19.9/10.7 vs �13.3/7.3mm Hg) butthat one could also expect a significantly greaterreduction in cerebral risk (�63 vs �49%) andcoronary disease (�46 vs �34%). For example, thestandard doses are 2.5mg for bendroflumethiazide,50mg for atenolol, 10mg for lisinopril, 80mg forvalsartan and 5mg for amlodipine.

In accordance with the work of Dickerson et al.,the third active principle to add in the event of aninsufficient response to a RAS inhibitorþdiureticcombination can only be a calcium channel blocker.

Curiously, the doctors in our study rarely pre-scribed a third drug. They may not have wished toincrease the number of different drug doses ormaybe they had previous experience of poortolerance to a calcium channel blocker.

Finally, the absence of any change in treatmentwas seen in a significant proportion of cases in ourstudy, especially in patients receiving the dualtherapy at maximal doses. Thus, the doctors settlefor BP values that are higher than the objective,despite the fact that the advantages of achievingthese objectives have now been established.32,33 Thisinertia is justifiably criticized in the literature.34 Themain reason for not modifying the treatment istolerability. Good tolerability is indeed essential toreduce the risk of intentional poor adherence35 and,in the same proportion, condition patient satisfac-tion for correct BP control, as demonstrated in anAmerican survey in patients contacted by e-mail.36

Doctors also want to try and limit the number ofdoses; indeed the risk of decreased adherence dueto an increase in the number of daily doses is wellrecognized.37

Blood pressure measurement outside the officeThe numerous sources of error associated with theclinical measurement of BP justify self-measure orABPM in the event of an apparently poor responseto antihypertensive treatment.

The SHEAF study38 revealed that discrepanciesbetween self-measurement and clinical measure-ment data are common: 14% of patients who werenormalized by the treatment according to the first

method were not according to the second method,thus defining ‘white coat HT’, whereas 9% were inthe opposite situation, characteristic of ‘masked’ HT.This study also demonstrated that the prognosis forthese masked hypertensives was even poorer thanthat of patients considered as poorly controlled withboth methods.

We specifically questioned our practitioners aboutABPM, whose contribution has also been welldocumented in apparently resistant or non-respon-sive hypertensives: in the study of Brown et al.,39

nearly one-quarter of patients sent to specializedclinics for poorly controlled BP despite receivingthree drugs actually had an ABPM during activity ofo135/85mm Hg. In common with self-measure-ment, the prognostic value of ambulatory BP is nowwell documented.40,41

European guidelines specify that informationprovided by ABPM should not replace informationfrom clinical measurements, but should be consid-ered as complementary, given that clinical BP andambulatory BP over 24h are only roughly corre-lated.42 The JNC 7 retains indications for ABPM,including the evaluation of hypertension in caseswith no target organ damage (to demonstrate apossible white-coat effect), and to investigate pa-tients with an apparently resistant hypertension.According to recent French guidelines,43 out ofoffice BP measurements are recommended to beperformed in all elderly patients as well as inpatients with mild-to-moderate hypertension andwithout target organ damage.

In our study, the implementation rate of ABPM inthe last 12 months or planned in the near futureappeared relatively high as it was prescribed in oneout of four patients. This was independent of thedose of the dual therapy, but was however morecommon in younger patients, in those with higherBP, with SD hypertension, with recently instigateddual therapy (less than 3 months) and in thosewith associated risk factors and cardiovascularpathologies.

Conclusion

The findings of this study show that GP do notfavour prescription of a third active principle inpatients who are not normalized by a fixed-dosetherapy combining a RAS inhibitor and a diuretic,despite the recognized efficacy of this strategy. Theyprefer to change the treatment: that is, to increasedosage and/or to switch to another class or withinthe same class. The choice of a treatment that isoften too weak can be explained by poor knowledgeof BP objectives and of the demonstrated benefits ofachieving these targets on cardiovascular morbidityand mortality. It can also be explained by the desireto maintain good tolerability and limit the numberof doses to maintain good patient adherence.Ambulatory blood pressure monitoring is employed


416


in one-quarter of cases, a proportion that is satisfac-tory, but which could still be improved.

The new French guidelines for hypertensionmanagement43 provide clinicians with updatedrecommendations regarding therapeutic strategies.Their widespread implementation should reason-ably result in minimizing the rate of suboptimaltreatment in hypertensive patients.

Acknowledgements

We are indebted to Simon Jarman for his linguisticassistance.

References

1 Girerd X, Mourad JJ, Vaisse B, Poncelet P, Mallion JM,Herpin D. Estimation du nombre des sujets traites pourune HTA, un diabete ou une hyperlipidemie en France:etude FLAHS 2002. Arch Mal Coeur 2003; 96: 750–753.

2 Frerot L, Le Fur P, Le Pape A, Sermet C. L’hypertensionarterielle en France: prevalence et prise en chargetherapeutique. Bulletin d’information en economie dela sante. Questions d’economie de la sante. CREDESSeptembre 1999; 22: 1–9.

3 Chamontin B, Poggi L, Lang T, Menard J, Chevalier H,Gallois H et al. Prevalence, treatment, and control ofhypertension in the French population: data from asurvey on high blood pressure in general practice,1994. Am J Hypertens 1998; 11: 759–762.

4 Mounier-Vehier C, Amah G, Covillard J, Jaboureck O,Phan TM. Prise en charge de l’HTA essentielle et duniveau de risque cardiovasculaire. Observatoire enmedecine generale: enquete nationale PHENOMEN.Arch Mal Coeur Vaiss 2002; 95: 667–672.

5 Amar J, Vaur L, Perret M, Bailleau C, Etienne S,Chamontin B et al. Hypertension in high-risk patients:beware of the underuse of effective combinationtherapy (results of the PRATIK study). J Hypertens2002; 20: 779–784.

6 Agence Nationale d’Accreditation et d’Evaluation enSante (ANAES). Prise en Charge des Patients AdultesAtteints d’hypertension Arterielle Essentielle. ANAES:Paris, 2000.

7 Tilly B, Guilhot J, Slanave B, Garrigue P, Fender P,Allemand H. Enquete nationale de l’assurance maladiesur la prise en charge medicale de l’hypertensionarterielle severe exoneree du ticket moderateur enFrance en 1999. Program Natl Sante Publique 2000,1–5.

8 Heart Outcomes Prevention Evaluation Study Investi-gators. Effects of ramipril on cardiovascular andmicrovascular outcomes in people with diabetesmellitus: results of the HOPE study and MICRO-HOPEsubstudy. Lancet 2000; 355: 253–259.

9 Hansson L, Lindholm LH, Niskanen L, Lanke J, HednerT, Niklason A et al. Effect of angiotensin-converting-enzyme inhibition compared with conventionaltherapy on cardiovascular morbidity and mortalityin hypertension: the Captopril Prevention Project(CAPPP) randomized trial. Lancet 1999; 353:611–616.

10 Nelson MR, Reid CM, Krum H, Ryan P, Wing LM,McNeil JJ et al. Management Committee, SecondAustralian National Blood Pressure Study. Short-termpredictors of maintenance of normotension after with-drawal of antihypertensive drugs in the secondAustralian National Blood Pressure Study (ANBP2).Am J Hypertens 2003; 16: 39–45.

11 Dahlof B, Devereux RB, Kjeldsen SE, Julius S, BeeversG, de Faire U et al. Cardiovascular morbidity andmortality in the Losartan Intervention For Endpointreduction in hypertension study (LIFE): a randomisedtrial against atenolol. Lancet 2002; 359: 995–1003.

12 Julius S, Kjeldsen SE, Weber M, Brunner HR, Ekman S,Hansson L et al. Outcomes in hypertensive patients athigh cardiovascular risk treated with regimens basedon valsartan or amlodipine: the VALUE randomisedtrial. Lancet 2004; 363: 2022–2031.

13 Mallion JM, Asmar R, Poggi L, Safar M, Andre JL,Balansard P et al. Pression arterielle – Automesure.Recommandations. Societe Francaise d’HypertensionArterielle – Groupe de la Mesure. Arch Mal CoeurVaiss 1989; 82: 1001–1005.

14 Marques-Vidal P, Arveiler D, Amouyel P, Bingham A,Ferrieres J. Sex differences in awareness andcontrol of hypertension in France. J Hypertens 1997;15: 1205–1210.

15 Lang T, de Gaudemaris R, Chatellier G, Hamici L,Diene E, Epidemiology Group of the Societe Francaised’Hypertension Arterielle and the IHPAF WorkingPhysicians Group. Prevalence and therapeutic controlof hypertension in 30 000 subjects in the workplace.Hypertension 2001; 38: 449–454.

16 Colhoun HM, Dong W, Poulter NR. Blood pressurescreening, management and control in England: resultsfrom the health survey for England 1994. J Hypertens1998; 16: 747–752.

17 Smith WC, Lee AJ, Crombie IK, Tunstall-Pedoe H.Control of blood pressure in Scotland: the rule ofhalves. BMJ 1990; 300: 981–983.

18 Hajjar I, Kotchen TA. Trends in prevalence, awareness,treatment, and control of hypertension in the UnitedStates, 1988–2000. JAMA 2003; 290: 199–206.

19 Ragot S, Sosner P, Herpin D. HTA non controlees:interet pharmaco-economique d’un bilan hospitalier.Arch Mal Coeur Vaiss 2002; 95: 733–737.

What is known on this topicK A high proportion of hypertensives are poorly controlled

despite their treatmentK The under-use of combination therapies and the suboptimal

utilization of triple therapiesK The lack of standardization of the therapeutic strategies for the

management of patients who are resistant to dual therapyK ABPM contribution is well documented in apparently resistant

or non-responsive hypertensives

What this study addsK General practitioners prefer to switch to a new fixed-dose or

separate combination rather than add a third drug in patientspoorly controlled with a fixed-dose renin–angiotensin systeminhibitor and diuretic combination

K This therapeutic decision was more common in the patientswith the highest BP, those with systolic–diastolichypertension, those whose dual therapy included an ACEI,those whose treatment was longstanding

K The implementation rate of ABPM in these poorly controlledpatients appeared relatively high as it was prescribed in oneout of four patients


417


20 Asmar R, Vol S, Pannier B, Brisac AM, Tichet J, ElHasnaoui A. High blood pressure and associatedcardiovascular risk factors in France. J Hypertens2001; 19: 1727–1732.

21 Law MR, Wald NJ, Morris JK, Jordan RE. Value of lowdose combination treatment with blood pressure low-ering drugs: analysis of 354 randomised trials. BMJ2003; 326: 1427–1434.

22 Morgan TO, Anderson AI, MacInnis RJ. ACE inhibi-tors, beta-blockers, calcium blockers, and diuretics forthe control of systolic hypertension. Am J Hypertens2001; 14: 241–247.

23 Magee PF, Freis ED. Is low-dose hydrochlorothiazideeffective? Hypertension 1986; 8: 135–139.

24 Jounela AJ, Lilja M, Lumme J, Morlin C, Hoyem A,Wessel-Aas T et al. Relation between low dose ofhydrochlorothiazide, antihypertensive effect andadverse effects. Blood Press 1994; 3: 231–235.

25 Girerd X, Denolle T, Yau C, Fiquet B, Brunel P, MoulinB et al. Automated office and home phone-transmittedblood pressure recordings in uncontrolled hyperten-sion treated with valsartan and hydrochlorothiazide.Blood Press 2004; S2: 18–24.

26 Dickerson JE, Hingorani AD, Ashby MJ, Palmer CR,Brown MJ. Optimisation of antihypertensive treatmentby crossover rotation of four major classes. Lancet1999; 353: 2008–2013.

27 Williams B, Poulter NR, Brown MJ, Davis M, McInnesGT, Potter JF et al. British Hypertension Societyguidelines for hypertension management 2004 (BHS-IV): summary. BMJ 2004; 328: 634–640.

28 Bloom BS. Continuation of initial antihypertensivemedication after 1 year of therapy. Clin Ther 1998; 20:671–681.

29 Bourgault C, Senecal M, Brisson M, Marentette MA,Gregoire JP. Persistence and discontinuation patternsof antihypertensive therapy among newly treatedpatients: a population-based study. J Hum Hypertens2005; 19: 607–613.

30 Chou CC, Lee MS, Ke CH, Chuang MH. Factorsinfluencing the switch in the use of antihypertensivemedications. Int J Clin Pract 2005; 59: 85–91.

31 Borenstein J, Whyte JL, Badamgarav E, Vogel D,Deutsch S, Weingarten S et al. Physician practicepatterns in the treatment of isolated systolic hyperten-sion in a primary care setting. J Clin Hypertens(Greenwich) 2002; 4: 93–100.

32 Hansson L, Zanchetti A, Carruthers SG, Dahlof B,Elmfeldt D, Julius S et al. Effects of intensive blood-pressure lowering and low-dose aspirin in patientswith hypertension: principal results of the Hyperten-sion Optimal Treatment (HOT) randomised trial. HOTStudy Group. Lancet 1998; 351: 1755–1762.

33 UK Prospective Diabetes Study Group. Cost effective-ness analysis of improved blood pressure control inhypertensive patients with type 2 diabetes: UKPDS 40.BMJ 1998; 317: 720–726.

34 Phillips LS, Branch WT, Cook CB, Doyle JP, El-KebbiIM, Gallina Dl et al. Clinical inertia. Ann Intern Med2001; 135: 825–834.

35 Toyoshima H, Takahashi K, Akera T. The impact ofside effects on hypertension management: a Japanesesurvey. Clin Ther 1997; 19: 1458–1469.

36 Chen K, Chiou CF, Plauschinat CA, Frech F, Harper A,Dubois R. Patient satisfaction with antihypertensivetherapy. J Hum Hypertens 2005; 19: 793–799.

37 Eisen SA, Miller DK, Woodward RS, Spitznagel E,Przybeck TR. The effect of prescribed daily dosefrequency on patient medication compliance. ArchIntern Med 1990; 150: 1881–1884.

38 Mallion JM, Genes N, Vaur L, Clerson P, Vaisse B,Bobrie G et al. Blood pressure levels, risk factors andantihypertensive treatments: lessons from the SHEAFstudy. J Hum Hypertens 2001; 15: 841–848.

39 Brown MA, Buddle ML, Martin A. Is resistanthypertension really resistant? Am J Hypertens 2001;14: 1263–1269.

40 Verdecchia P, Porcellati C, Schillaci G, Borgioni C,Ciucci A, Battistelli M et al. Ambulatory blood pressure.An independent predictor of prognosis in essentialhypertension. Hypertension 1994; 24: 793–801.

41 Clement DL, De Buyzere ML, De Bacquer DA, deLeeuw PW, Duprez DA, Fagard RH et al. Office versusAmbulatory Pressure Study Investigators. Prognosticvalue of ambulatory blood-pressure recordings inpatients with treated hypertension. N Engl J Med2003; 348: 2407–2415.

42 Mancia G, Parati G. Ambulatory blood pressuremonitoring and organ damage. Hypertension 2000;36: 894–900.

43 Haute Autorite de Sante (HAS). Prise en Charge desPatients Adultes Atteints d’hypertension ArterielleEssentielle – Actualisation 2005. HAS: Paris, 2005(www.has-sante.fr).


418


original article - Nature

Documents

Transcript of original article - Nature