Evaluation of Adherence Should Become an Integral Part of Assessment of Patients With Apparently Treatment-Resistant Hypertension
2016; Lippincott Williams & Wilkins; Volume: 68; Issue: 2 Linguagem: Inglês
10.1161/hypertensionaha.116.07464
ISSN1524-4563
AutoresE. Berra, Michel Azizi, Arnaud Capron, Aud Høieggen, Franco Rabbia, Sverre E. Kjeldsen, Jan A. Staessen, Pierre Wallemacq, Alexandre Persu,
Tópico(s)Hormonal Regulation and Hypertension
ResumoHomeHypertensionVol. 68, No. 2Evaluation of Adherence Should Become an Integral Part of Assessment of Patients With Apparently Treatment-Resistant Hypertension Free AccessResearch ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyRedditDiggEmail Jump toSupplementary MaterialsFree AccessResearch ArticlePDF/EPUBEvaluation of Adherence Should Become an Integral Part of Assessment of Patients With Apparently Treatment-Resistant Hypertension Elena Berra, Michel Azizi, Arnaud Capron, Aud Høieggen, Franco Rabbia, Sverre E. Kjeldsen, Jan A. Staessen, Pierre Wallemacq, and Alexandre Persu Elena BerraElena Berra From the Pole of Cardiovascular Research, Institut de Recherche Expérimentale et Clinique, Université Catholique de Louvain, Brussels, Belgium (A.P., E.B.); Department of Medical Sciences, Internal Medicine and Hypertension Division, AOU Città della Salute e della Scienza, Turin, Italy (F.R., E.B.); Division of Cardiology, Cliniques Universitaires Saint-Luc, Université Catholique de Louvain, Brussels, Belgium (A.P.); Studies Coordinating Centre, Research Unit Hypertension and Cardiovascular Epidemiology, KU Leuven Department of Cardiovascular Sciences, University of Leuven, Leuven, Belgium (J.A.S.); R&D VitaK Group, Maastricht University, Maastricht, The Netherlands (J.A.S.); Paris-Descartes University, F-75005 Paris, France (M.A.); Assistance Publique-Hôpitaux de Paris, Hypertension Unit, Hôpital Européen Georges Pompidou, F-75015 Paris, France (M.A.); Institut National de la Santé et de la Recherche Médicale (INSERM), Centre d'Investigations Cliniques 1418, F-75015 Paris, France (M.A.); Department of Cardiology (S.E.K.) and Department of Nephrology (A.H.), Ullevaal University Hospital, University of Oslo, Oslo, Norway; and Clinical Chemistry Department, Cliniques Universitaires St Luc, Brussels, Belgium; Louvain Center for Toxicology and Applied Pharmacology, Université Catholique de Louvain, Brussels, Belgium (P.W, A.C.). , Michel AziziMichel Azizi From the Pole of Cardiovascular Research, Institut de Recherche Expérimentale et Clinique, Université Catholique de Louvain, Brussels, Belgium (A.P., E.B.); Department of Medical Sciences, Internal Medicine and Hypertension Division, AOU Città della Salute e della Scienza, Turin, Italy (F.R., E.B.); Division of Cardiology, Cliniques Universitaires Saint-Luc, Université Catholique de Louvain, Brussels, Belgium (A.P.); Studies Coordinating Centre, Research Unit Hypertension and Cardiovascular Epidemiology, KU Leuven Department of Cardiovascular Sciences, University of Leuven, Leuven, Belgium (J.A.S.); R&D VitaK Group, Maastricht University, Maastricht, The Netherlands (J.A.S.); Paris-Descartes University, F-75005 Paris, France (M.A.); Assistance Publique-Hôpitaux de Paris, Hypertension Unit, Hôpital Européen Georges Pompidou, F-75015 Paris, France (M.A.); Institut National de la Santé et de la Recherche Médicale (INSERM), Centre d'Investigations Cliniques 1418, F-75015 Paris, France (M.A.); Department of Cardiology (S.E.K.) and Department of Nephrology (A.H.), Ullevaal University Hospital, University of Oslo, Oslo, Norway; and Clinical Chemistry Department, Cliniques Universitaires St Luc, Brussels, Belgium; Louvain Center for Toxicology and Applied Pharmacology, Université Catholique de Louvain, Brussels, Belgium (P.W, A.C.). , Arnaud CapronArnaud Capron From the Pole of Cardiovascular Research, Institut de Recherche Expérimentale et Clinique, Université Catholique de Louvain, Brussels, Belgium (A.P., E.B.); Department of Medical Sciences, Internal Medicine and Hypertension Division, AOU Città della Salute e della Scienza, Turin, Italy (F.R., E.B.); Division of Cardiology, Cliniques Universitaires Saint-Luc, Université Catholique de Louvain, Brussels, Belgium (A.P.); Studies Coordinating Centre, Research Unit Hypertension and Cardiovascular Epidemiology, KU Leuven Department of Cardiovascular Sciences, University of Leuven, Leuven, Belgium (J.A.S.); R&D VitaK Group, Maastricht University, Maastricht, The Netherlands (J.A.S.); Paris-Descartes University, F-75005 Paris, France (M.A.); Assistance Publique-Hôpitaux de Paris, Hypertension Unit, Hôpital Européen Georges Pompidou, F-75015 Paris, France (M.A.); Institut National de la Santé et de la Recherche Médicale (INSERM), Centre d'Investigations Cliniques 1418, F-75015 Paris, France (M.A.); Department of Cardiology (S.E.K.) and Department of Nephrology (A.H.), Ullevaal University Hospital, University of Oslo, Oslo, Norway; and Clinical Chemistry Department, Cliniques Universitaires St Luc, Brussels, Belgium; Louvain Center for Toxicology and Applied Pharmacology, Université Catholique de Louvain, Brussels, Belgium (P.W, A.C.). , Aud HøieggenAud Høieggen From the Pole of Cardiovascular Research, Institut de Recherche Expérimentale et Clinique, Université Catholique de Louvain, Brussels, Belgium (A.P., E.B.); Department of Medical Sciences, Internal Medicine and Hypertension Division, AOU Città della Salute e della Scienza, Turin, Italy (F.R., E.B.); Division of Cardiology, Cliniques Universitaires Saint-Luc, Université Catholique de Louvain, Brussels, Belgium (A.P.); Studies Coordinating Centre, Research Unit Hypertension and Cardiovascular Epidemiology, KU Leuven Department of Cardiovascular Sciences, University of Leuven, Leuven, Belgium (J.A.S.); R&D VitaK Group, Maastricht University, Maastricht, The Netherlands (J.A.S.); Paris-Descartes University, F-75005 Paris, France (M.A.); Assistance Publique-Hôpitaux de Paris, Hypertension Unit, Hôpital Européen Georges Pompidou, F-75015 Paris, France (M.A.); Institut National de la Santé et de la Recherche Médicale (INSERM), Centre d'Investigations Cliniques 1418, F-75015 Paris, France (M.A.); Department of Cardiology (S.E.K.) and Department of Nephrology (A.H.), Ullevaal University Hospital, University of Oslo, Oslo, Norway; and Clinical Chemistry Department, Cliniques Universitaires St Luc, Brussels, Belgium; Louvain Center for Toxicology and Applied Pharmacology, Université Catholique de Louvain, Brussels, Belgium (P.W, A.C.). , Franco RabbiaFranco Rabbia From the Pole of Cardiovascular Research, Institut de Recherche Expérimentale et Clinique, Université Catholique de Louvain, Brussels, Belgium (A.P., E.B.); Department of Medical Sciences, Internal Medicine and Hypertension Division, AOU Città della Salute e della Scienza, Turin, Italy (F.R., E.B.); Division of Cardiology, Cliniques Universitaires Saint-Luc, Université Catholique de Louvain, Brussels, Belgium (A.P.); Studies Coordinating Centre, Research Unit Hypertension and Cardiovascular Epidemiology, KU Leuven Department of Cardiovascular Sciences, University of Leuven, Leuven, Belgium (J.A.S.); R&D VitaK Group, Maastricht University, Maastricht, The Netherlands (J.A.S.); Paris-Descartes University, F-75005 Paris, France (M.A.); Assistance Publique-Hôpitaux de Paris, Hypertension Unit, Hôpital Européen Georges Pompidou, F-75015 Paris, France (M.A.); Institut National de la Santé et de la Recherche Médicale (INSERM), Centre d'Investigations Cliniques 1418, F-75015 Paris, France (M.A.); Department of Cardiology (S.E.K.) and Department of Nephrology (A.H.), Ullevaal University Hospital, University of Oslo, Oslo, Norway; and Clinical Chemistry Department, Cliniques Universitaires St Luc, Brussels, Belgium; Louvain Center for Toxicology and Applied Pharmacology, Université Catholique de Louvain, Brussels, Belgium (P.W, A.C.). , Sverre E. KjeldsenSverre E. Kjeldsen From the Pole of Cardiovascular Research, Institut de Recherche Expérimentale et Clinique, Université Catholique de Louvain, Brussels, Belgium (A.P., E.B.); Department of Medical Sciences, Internal Medicine and Hypertension Division, AOU Città della Salute e della Scienza, Turin, Italy (F.R., E.B.); Division of Cardiology, Cliniques Universitaires Saint-Luc, Université Catholique de Louvain, Brussels, Belgium (A.P.); Studies Coordinating Centre, Research Unit Hypertension and Cardiovascular Epidemiology, KU Leuven Department of Cardiovascular Sciences, University of Leuven, Leuven, Belgium (J.A.S.); R&D VitaK Group, Maastricht University, Maastricht, The Netherlands (J.A.S.); Paris-Descartes University, F-75005 Paris, France (M.A.); Assistance Publique-Hôpitaux de Paris, Hypertension Unit, Hôpital Européen Georges Pompidou, F-75015 Paris, France (M.A.); Institut National de la Santé et de la Recherche Médicale (INSERM), Centre d'Investigations Cliniques 1418, F-75015 Paris, France (M.A.); Department of Cardiology (S.E.K.) and Department of Nephrology (A.H.), Ullevaal University Hospital, University of Oslo, Oslo, Norway; and Clinical Chemistry Department, Cliniques Universitaires St Luc, Brussels, Belgium; Louvain Center for Toxicology and Applied Pharmacology, Université Catholique de Louvain, Brussels, Belgium (P.W, A.C.). , Jan A. StaessenJan A. Staessen From the Pole of Cardiovascular Research, Institut de Recherche Expérimentale et Clinique, Université Catholique de Louvain, Brussels, Belgium (A.P., E.B.); Department of Medical Sciences, Internal Medicine and Hypertension Division, AOU Città della Salute e della Scienza, Turin, Italy (F.R., E.B.); Division of Cardiology, Cliniques Universitaires Saint-Luc, Université Catholique de Louvain, Brussels, Belgium (A.P.); Studies Coordinating Centre, Research Unit Hypertension and Cardiovascular Epidemiology, KU Leuven Department of Cardiovascular Sciences, University of Leuven, Leuven, Belgium (J.A.S.); R&D VitaK Group, Maastricht University, Maastricht, The Netherlands (J.A.S.); Paris-Descartes University, F-75005 Paris, France (M.A.); Assistance Publique-Hôpitaux de Paris, Hypertension Unit, Hôpital Européen Georges Pompidou, F-75015 Paris, France (M.A.); Institut National de la Santé et de la Recherche Médicale (INSERM), Centre d'Investigations Cliniques 1418, F-75015 Paris, France (M.A.); Department of Cardiology (S.E.K.) and Department of Nephrology (A.H.), Ullevaal University Hospital, University of Oslo, Oslo, Norway; and Clinical Chemistry Department, Cliniques Universitaires St Luc, Brussels, Belgium; Louvain Center for Toxicology and Applied Pharmacology, Université Catholique de Louvain, Brussels, Belgium (P.W, A.C.). , Pierre WallemacqPierre Wallemacq From the Pole of Cardiovascular Research, Institut de Recherche Expérimentale et Clinique, Université Catholique de Louvain, Brussels, Belgium (A.P., E.B.); Department of Medical Sciences, Internal Medicine and Hypertension Division, AOU Città della Salute e della Scienza, Turin, Italy (F.R., E.B.); Division of Cardiology, Cliniques Universitaires Saint-Luc, Université Catholique de Louvain, Brussels, Belgium (A.P.); Studies Coordinating Centre, Research Unit Hypertension and Cardiovascular Epidemiology, KU Leuven Department of Cardiovascular Sciences, University of Leuven, Leuven, Belgium (J.A.S.); R&D VitaK Group, Maastricht University, Maastricht, The Netherlands (J.A.S.); Paris-Descartes University, F-75005 Paris, France (M.A.); Assistance Publique-Hôpitaux de Paris, Hypertension Unit, Hôpital Européen Georges Pompidou, F-75015 Paris, France (M.A.); Institut National de la Santé et de la Recherche Médicale (INSERM), Centre d'Investigations Cliniques 1418, F-75015 Paris, France (M.A.); Department of Cardiology (S.E.K.) and Department of Nephrology (A.H.), Ullevaal University Hospital, University of Oslo, Oslo, Norway; and Clinical Chemistry Department, Cliniques Universitaires St Luc, Brussels, Belgium; Louvain Center for Toxicology and Applied Pharmacology, Université Catholique de Louvain, Brussels, Belgium (P.W, A.C.). , and Alexandre PersuAlexandre Persu From the Pole of Cardiovascular Research, Institut de Recherche Expérimentale et Clinique, Université Catholique de Louvain, Brussels, Belgium (A.P., E.B.); Department of Medical Sciences, Internal Medicine and Hypertension Division, AOU Città della Salute e della Scienza, Turin, Italy (F.R., E.B.); Division of Cardiology, Cliniques Universitaires Saint-Luc, Université Catholique de Louvain, Brussels, Belgium (A.P.); Studies Coordinating Centre, Research Unit Hypertension and Cardiovascular Epidemiology, KU Leuven Department of Cardiovascular Sciences, University of Leuven, Leuven, Belgium (J.A.S.); R&D VitaK Group, Maastricht University, Maastricht, The Netherlands (J.A.S.); Paris-Descartes University, F-75005 Paris, France (M.A.); Assistance Publique-Hôpitaux de Paris, Hypertension Unit, Hôpital Européen Georges Pompidou, F-75015 Paris, France (M.A.); Institut National de la Santé et de la Recherche Médicale (INSERM), Centre d'Investigations Cliniques 1418, F-75015 Paris, France (M.A.); Department of Cardiology (S.E.K.) and Department of Nephrology (A.H.), Ullevaal University Hospital, University of Oslo, Oslo, Norway; and Clinical Chemistry Department, Cliniques Universitaires St Luc, Brussels, Belgium; Louvain Center for Toxicology and Applied Pharmacology, Université Catholique de Louvain, Brussels, Belgium (P.W, A.C.). Originally published13 Jun 2016https://doi.org/10.1161/HYPERTENSIONAHA.116.07464Hypertension. 2016;68:297–306Other version(s) of this articleYou are viewing the most recent version of this article. Previous versions: January 1, 2016: Previous Version 1 Renal Denervation: An Eye OpenerSince the publication of the first Symplicity studies in 2009 to 2010, renal sympathetic denervation gained acceptance as a novel treatment of drug-resistant hypertension. The latter has been defined as a blood pressure (BP) >140/90 mm Hg, despite appropriate lifestyle measures plus a diuretic and 2 other antihypertensive drugs belonging to different classes at adequate doses.1 According to the US definition, patients with controlled BP on ≥4 antihypertensive drugs are also considered as resistant hypertensives.2 However, a substantial proportion of patients with apparently resistant hypertension are in fact poorly adherents to drug treatment. The highly variable BP response to renal denervation (RDN)3–5 prompted to a more rigorous evaluation of eligible patients, with the goal to exclude false resistant hypertension, because of poor adherence to drug treatment.6–8 In particular, several publications documented a high proportion of low drug adherence in patients with apparently resistant hypertension (23%–66%), using witnessed drug intake9 or plasma/urine drug determinations10–18 (Figure 1).Download figureDownload PowerPointFigure 1. Proportion of poor or nonadherence according to drug monitoring in different cohorts of patients with apparently resistant hypertension. Black indicates total nonadherence, whereas gray indicates partial adherence. Partial adherence was defined as the presence of at least one undetectable drug10–12,14,16–18 or as the presence of fewer medications than prescribed.13,15Furthermore, RDN studies shed the light on the dynamic character of drug adherence. Inclusion in RDN trials may influence drug adherence in various, unpredictable directions.6 In some patients, close follow-up and massive attention devoted to them may lead to improved adherence to lifestyle measures and drug treatment, particularly in the RDN arm (Hawthorne effect). Other patients may stop their medications after RDN according to their perception that the intervention cured their hypertension. On the other hand, patients from the control group may have not taken their medications properly to keep their BP high, in the hope that this will make them eligible for crossover to the RDN group.Overall, RDN trials confirm (1) that poor drug adherence is a frequent cause of apparently resistant and difficult- to-treat hypertension, (2) that drug adherence is a dynamic phenomenon influenced by complex psychosocial determinants and cannot be captured by any single assessment, (3) and that changes in drug adherence are a major potential confounder in trials assessing new treatment modalities of resistant hypertension.Although the prevalence of resistant hypertension has been estimated to be 10% to 30%,19,20 it may decrease to 80%) baseline drug adherence, in terms of both BP control and regression of target organ damage.22 Therefore, from a public health and pharmacoeconomic perspective, diagnosis and management of poor drug adherence in patients with apparently resistant hypertension23 is a priority.Identification of poorly adherent patients among those with apparently resistant hypertension will avoid unnecessary and potentially harmful treatment intensification and allow implementation of strategies to improve drug adherence.14,24 This approach would be expected to result in a more cost-effective allocation of health resources. Unfortunately, healthcare providers underestimate the size of the problem of poor drug adherence. Assessment of adherence is not part of routine clinical practice and is seldom performed in studies, even those testing drugs or interventions in patients with resistant hypertension. The finding of persistent tachycardia in a patient on β-blockers suggests a poor adherence or may orient toward a secondary form of hypertension. However, intuitive assessment of adherence by physicians is specific (most patients identified as nonadherent by their physician are indeed so), but lacks sensitivity (24%–62% depending on the definition of adherence).25Moreover, unsubstantiated suspicion of poor adherence in difficult-to-treat patients with resistant hypertension contributes to medical skepticism and therapeutic inertia. Knowing which of their patients do not achieve BP control despite demonstrated good adherence, physicians would feel more motivated and responsible for the optimization of their antihypertensive regimen.The authors of this review article make a plea for a more widespread, systematic, and standardized use of assessment of drug adherence in patients with apparently resistant hypertension, both in clinical practice and for evaluation of novel drug and interventional strategies, with particular emphasis on drug monitoring its advantages and limitations.Drug Adherence in Resistant HypertensionDrug adherence is a major concern in hypertensive patients at large, and poor adherence has a demonstrated impact on cardiovascular prognosis (see online-only Data Supplement). Not unexpectedly, poor adherence is particularly frequent in apparently resistant hypertensive patients. Nine recent studies totalizing 747 patients from 5 countries10–18 evaluated the level of adherence in outpatients with resistant or difficult-to-control hypertension using high-performance liquid chromatography coupled with mass spectrometry in urine and plasma. Overall, clinical characteristics, BP values, and average of prescribed drugs per day were similar10–18 (Table 1). The percentage of poor and full nonadherence to drug regimens in these 9 studies ranged from 13% to 46% and from 2% to 35%, respectively (Figure 1).Table 1. Main Characteristics of Patients With Apparently Resistant Hypertension in Whom Drug Adherence Was Tested Using Drug MonitoringPatients' CharacteristicsCeral et al 201110Jung et al 201311Strauch et al 201312Tomaszewski et al 201413Brinker et al 201414Ewen et al 201515Patel et al 201516Hamdidouche et al 201517Schmieder et al 201618Resistant hypertensive+++++++++Number of patients8476163661756100248279Mean age, y5558545763NR6365NR60Male, %6058564247NR6762NR72BMI, kg/m2NRNR32.331.6*31.9*NR30.833.0NR31.1Number of drugs5.0±1.255.2±1.3NRNRNR5.2±1.43.3±1.7NR6.0Office BP, mm Hg172/97NR175/100168/95*168/94*168/94167/88172/92NR158/8824-h ambulatory BP, mm HgNRNR156/91160/91*160/89*NR154/86162/88NR155/88HPLC-MSbloodurinebloodurineurinebloodblood/ urineurineurineurineBMI indicates body mass index; BP, blood pressure; HPLC-MS, high-performance liquid chromatography coupled with mass spectrometry; and NR, not reported.*Not available for all patients.However, it should be kept in mind that these measurements give only a snapshot of evaluation of adherence. In view of the known phenomenon of white coat adherence or toothbrush effect (increased adherence during the few days preceding a medical contact),26 daily-life drug adherence between 2 visits may be overestimated. Unfortunately, most of these studies have important limitations: (1) no rigorous definition of resistant hypertension or of partial adherence was provided10,13,14; (2) details on the antihypertensive regimen were not reported13,14; (3) patients included were not always on optimal antihypertensive therapy and did not always meet strict criteria of drug resistance10,13,14; (4) details on methodology of drug dosage were not reported14,16; (5) the timing of the samplings after the last drug intake was not cleary defined13–16,18; (6) the possible influence of preanalytical errors and pharmacokinetics on the level of drugs detected was not discussed10,11,13,16,18; (7) it was not clearly explained whether patients gave informed consent to have their adherence assessed and when10,11,14; (8) and predictors of poor adherence were not analyzed.10,14–16,18Methods to Evaluate Drug AdherenceOver time, adherence has been studied using increasingly reliable and sophisticated methods. According to the method used, assessment of drug adherence can take place at different steps of the patient's pathway (physician's office, pharmacy, home; Figure 2). These different approaches to assess drug adherence can be divided into 2 categories: indirect and direct methods. Indirect methods include assessment by the clinician,25 self-assessment by the patient (interviews/questionnaires),27,28 measurement of pharmacodynamics parameters (heart rate for β-blockers, lack of rise of plasma renin for renin–angiotensin inhibitors, which may be also due to undetected primary aldosteronism-,29 as well as acetyl-SDKP measurements for angiotensin-converting enzyme inhibitors,),30 pill counts,31 and prescription refill. Direct methods include witnessed drug intake, Medication Event Monitoring System (MEMS), telemonitoring, and drug monitoring in body fluids.Download figureDownload PowerPointFigure 2. Patient pathway from drug prescription at the physician's office to next control visit, highlighting the different possible levels of drug adherence monitoring.The choice to use a method rather than another depends on multiple factors, including reliability, sensitivity to white-coat adherence, educational value, local facilities, long-term feasibility, patient profile, and financial resources (Table 2). All methods have limitations, and ideally, accurate evaluation of adherence should involve a combination of several approaches.Table 2. Advantages and Disadvantages of Different Methods to Assess Drug AdherenceAdvantages and DisadvantagesIndirect MethodsDirect MethodsAssessment by ClinicianQuestionnairesPill CountPrescription RefillWitnessed Drug IntakeElectronic MonitoringTele-MonitoringDrugs Dosage in Body FluidsAdvantagesObjectivity↓↓↓↓↓↓↓↓↓↓↓↓↑↑↑↑↑↑↑↑↑↑↑Accuracy↓↓↓↓↓↓↓↓↓↓↓↓↑↑↑↑↑↑↑↑↑↑↑↑Feasibility↑↑↑↑↑↑↑↑↑↑↑↑↓↓↓↑↓↓↓Educational value↓↓↓↑↑↑–↑↑↑↑↑↑↑↑↑↑DisadvantagesCost/workload↓↓↓↓↓↓↓↓↓↓↑↑↑↑↑↑↑↑↑↑↑↑White coat effect––↑↑↑––↓↓↓↑↑↑↑↑Social desirability bias↓↓↓↓↑↑↑↑↑↑↑↑↑–↑↑↑↑↑↓↓Manipulability↓↓↓↓↑↑↑↑↑↑↑↑↑↓↓↓↓↑↑↑↑↓↓↓↓↓Social desirability bias: people respond to questioning in ways that make them seem more appealing to others or to healthcare providers. Manipulability: patients willingly manipulate the results of the assessment.Bobrie et al evaluated adherence in a cohort of 164 resistant hypertensive patients on a standardized triple antihypertensive therapy, which were randomized to 2 different treatment strategies (sequential nephron blockade versus renin–angiotensin system blockade).32 Subsequently, the same group evaluated the influence of adherence on the efficacy of these 2 approaches.22 The authors used 2 indirect and 2 direct methods to assess adherence, assigning one point to each drug adherence measurement to calculate a global score.22 The superiority of sequential nephron blockade over renin–angiotensin system blockade in terms of BP control and regression of target organ damage (arterial stiffness and left ventricular mass) was significant only in adherent patients. Along similar lines, in a cohort of >240 000 newly treated hypertensive patients from the Italian Lombardy region, followed for 6 years, patients with a drug coverage ≥75% (24% of the cohort) had a 25% lower risk of cardiovascular events compared with those with a coverage <25% (22% of the cohort).33Indirect methods are simple, inexpensive, and time-efficient, and they imply a reasonable workload.27,34–36 On the other hand, the sensitivity of these methods is poor35,37–40; they are heavily dependent on patient behavior,41,42 affected by social desirability and recall biases,43–47 and fail to provide information about the timing of doses, which is an essential aspect of adherence.48 Accordingly, they are poorly correlated with direct methods such as MEMS and drug dosage in body fluids.35,37,42,44,45,47 In particular, in a sample of 47 patients with apparently resistant hypertension, poor adherence was grossly underestimated by the Morisky Medication Adherence Scale-8 (26%) compared with drug monitoring (51%).47 Compared with the latter, the sensitivity of Morisky Medication Adherence Scale-8 was found to be 26%, specificity 75%, positive predictive value 50%, and negative predictive value 51%.Direct methods are more accurate and reliable49 than indirect methods, but also more expensive and demanding in terms of human resources.44,50Witnessed drug intake combined with ambulatory BP measurement might appear as a gold standard but cannot give information about persistence. Its application requires a hospital environment and the involvement of dedicated, trained healthcare professionals. The strength of this method is to provide an immediate proof that the prescribed medication in the prescribed doses is effective, if taken. On the other hand, witnessed drug intake raises ethical issues. Its use is justified in highly contagious and potentially fatal diseases such as tuberculosis, but less in hypertension, which only in the long-term leads to cardiovascular complications.51 Therefore, this approach is appropriate only within the framework of clinical studies.9,52Telemonitoring allows obtaining a higher number of BP measures and improves the patient–physician communication. Furthermore, it improves patient compliance and hence BP control.53–55 Moreover, nowadays, the use of memory-equipped BP monitors allows overcoming problems related to the reliability of reports by patients.56 However, telemonitoring involves a high workload for the healthcare professionals, which limits its applicability in clinical practice.57,58The MEMS system provides detailed information about the timing of drug intake and missing doses.40,59 It can be used long term, favors patient empowerment, and has a demonstrated educational value.60,61 Accordingly, it has been associated with improved drug adherence.57,58,60–62 Nevertheless, electronic monitoring has an important limitation: it does not allow detecting patients who willingly open the box without swallowing the pill. Such behavior may be over-represented in the subgroup of difficult-to-treat patients referred for RDN.63,64 This intrinsic disadvantage can be overcome by the use of an ingestible sensor65,66 but the latter is expensive, is invasive, has not been validated on a large scale, and is not free from possible side effects, such as skin allergies caused by the patch and possible gut retention of its metal components.The alternative is measuring drug/metabolite levels in body fluids. In hypertensive patients, the most widely used matrices for drug measurements are blood and urine.10–18 High-performance liquid chromatography coupled with a sensitive detector such as mass spectrometry is considered the reference analytic technique.67 Although widely used for drug monitoring, blood has the disadvantage to require invasive sampling and relatively accurate blood drawing time, especially for drugs with higher clearance. Even though urine provides a larger detection window as compared with blood, none of these matrices are able to assess if the treatment is rigorously followed between medical visits. Possible alternative approaches and matrices for drug monitoring are discussed in the online-only Data Supplement.Drug monitoring is objective in detecting in a semiquantitative manner the presence or not of drugs in body fluids, but cannot give precise quantitative information, such as whether the patient is at his/her maximal tolerated dose.45,59 In contrast to electronic drug monitoring, it allows to test adherence for all drugs, which is not a trivial asset in view of the known differences in adherence according to drug class.68–70 Nevertheless, drug monitoring provides no information on the timing of drug intake or on the persistence on treatment. Furthermore, in contrast with MEMS,60,61 the educational value of repeated drug monitoring has not been studied in depth. Still, in a cohort of 56 patients with apparently resistant hypertension, Brinker et al documented a sustained BP decrease (−46±10/−26±14 mm Hg; P<0.01) after information about the undetectable level of prescribed drugs, focused discussion and counseling, in the absence of drug treatment adjustment.14Drug monitoring is sensitive to white-coat adherence, especially if the patient has been informed in advance.46 Nevertheless, this disadvantage can be partly overcome if drug monitoring is performed at short notice (ie, the day the patient provides his informed consent
Referência(s)