The aim of this study was to determine the clinical characteristics of patients with resistant hypertension (RH) and predictors among elderly Korean hypertensives. This prospective, multi-center, observational study evaluated 2439 elderly hypertensive patients between December 2008 and November 2011, who visited secondary hypertension clinics for high blood pressure (BP). Patients were categorized as resistant if their BP was ⩾140/90 mm Hg and if they reported using antihypertensive medications from three different drug classes, including a diuretic or drugs from ⩾4 antihypertensive drug classes, regardless of BP. Characteristics of patients with RH were compared with those of patients who were controlled with one or two antihypertensive medications after 6-month antihypertensive treatment. In comparison with 837 patients with non-RH, 404 patients with RH were more likely to be aware of their status of high BP before enrollment and have a high baseline systolic BP ⩾160 mm Hg, microalbuminuria, high body mass index (BMI) ⩾24 kg m−2 and diabetes mellitus (DM). In drug-naive patients, awareness of hypertension at baseline was the only independent predictor for RH. In elderly Korean hypertensives, BMI (⩾24 kg m−2), baseline systolic BP (⩾160 mm Hg), microalbuminuria, DM and awareness of hypertension showed an association with RH.
Inspite of active lifestyle modification and antihypertensive medication, in some patients, high blood pressure (BP) remains uncontrolled or requires >4 different classes of medications. Recently, considerable efforts have been made to classify these patients as resistant hypertension (RH). RH is defined as BP that remains above goal despite concurrent use of three antihypertensive agents of different classes, including a diuretic.1 RH also includes patients whose BP is controlled but require ⩾4 medications for the control of BP.1 Although the definition has been criticized for its ambiguity and drawbacks, it has been suggested to identify patients with secondary hypertension and those who benefit from specific treatments.
With regard to prevalence of RH, recent data showed that about 12.8% of the antihypertensive drug-treated population met the criteria for RH in the National Health and Nutrition Examination Survey from 2003 through 2008.2 Also, the other data indicated that its prevalence increased from 5.5% in 1988–1994 to 11.8% in 2005–2008 in all hypertensive patients.3 In terms of prognosis, over 3.8 years of median follow-up, patients with RH showed significantly higher cardiovascular event rates (unadjusted 18.0% vs 13.5%, P<0.001) compared with non-RH.4
With its unexpected high prevalence and poor prognosis,4, 5 the clinical importance of RH has been emphasized. The characteristics of RH were currently investigated in large observational data.5, 6, 7 However, most of these clinical trials were conducted in Western countries and their clinical implications in Asian patients are unknown.
In this study, we assessed the clinical characteristics and predictive factors of elderly patients with RH by comparing them with hypertensive patients controlled with one or two antihypertensive medications.
Materials and Methods
The HIT (Hypertension In elderly patienTs) registry was designed for the identification of the characteristics of elderly Korean hypertensives and for the assessment of antihypertensive treatment outcomes. This study is a prospective, multi-center, observational registry in which 43 hospitals in Korea participated. We enrolled consecutive patients who visited secondary hypertension clinics for high BP and were ⩾60 years of age, regardless of past medical history of hypertension. Patients participating in other clinical trials were excluded. At the baseline visit, demographic and anthropometric characteristics, awareness of hypertension, past antihypertensive treatment, past medical history (hyperlipidemia, diabetes mellitus (DM), family history of cardiovascular disease, physical inactivity and smoking) and target organ damage (microalbuminuria, heart disease, cerebrovascular accident, chronic kidney disease, peripheral artery disease and retinopathy) were recorded for all eligible patients under the guidance of attending physicians by using a preformed data collection format. Medications for antihypertensive treatment were prescribed at the physician’s discretion. Adherence to medications was checked out by the physicians at each visit. Patients were followed up to 6 months. BP measurements and dose titrations or adjustments of antihypertensive drugs were performed aggressively at intervals of 1 month. Drug side effects and orthostatic hypotension were reported at each visit.
A patient was classified as resistant if his or her BP was ⩾140/90 mm Hg despite concurrent use of antihypertensive medications from three different drug classes or drugs from ⩾4 antihypertensive drug classes, regardless of BP. One of the three or four agents should be a diuretic, and all agents should be prescribed at optimal dose amounts.1 Patient’s status of RH was evaluated at the 6-month (± 8 weeks) visit. Clinical characteristics of patients with RH were compared with those of patients who were controlled with one or two antihypertensive medications (non-RH). Predictive factors associated with RH were analyzed.
Measurements of both systolic BP (SBP) and diastolic BP were taken from both the arms two or three times at intervals of 2 min in a sitting position, using an automated oscillometric BP device (HEM 737, Omron Healthcare, Kyoto, Japan)8 in accordance with official recommendations for the measurement of BP.9
The local ethics committee at each hospital approved the use of clinical data for this study, and all patients provided written informed consent.
Definitions of variables
As previously noted, RH was defined as BP that remains above 140/90 mm Hg despite concurrent use of three antihypertensive medications, including one diuretic, or use of >4 antihypertensive medications, regardless of BP.1
Awareness of hypertension was determined by hypertensive patients responding to the question ‘Have you ever been told by a doctor or other healthcare professional that you had high blood pressure?’ Patients were considered to have family history of early cardiovascular disease if they had first-degree relatives who had cardiovascular disease when the age of men was <55 years and that of women was <65 years. Patients were regarded as being physically inactive when they did not perform regular aerobic exercise for at least 30 min more than 5 days per week.
Heart disease in target organ damage was defined as the presence of left ventricular hypertrophy on electrocardiogram, history of angina pectoris, myocardial infarction and coronary artery revascularization or heart failure. Cerebrovascular accident was defined as the history of cerebral infarct, cerebral hemorrhage or transient ischemic attack. Microalbuminuria was screened using MICRAL-TEST (F. Hoffmann-La Roche Ltd., Basel, Switzerland), which was confirmed by color change of the test stick. Chronic kidney disease was determined if estimated glomerular filtration rate was <60 ml per minper 1.73 m2 or proteinuria was>300 mg daily or >300 mg albumin/creatinine in spot urine. Glomerular filtration rate was estimated by the Modification of Diet in Renal Disease Study equation.10 Peripheral artery disease was determined if a patient had intermittent claudication or his/her ankle-brachial BP index was <0.9. Retinopathy was defined as retinal hemorrhage, exudate or papilledema upon funduscopic examination.
Continuous variables were expressed as mean±s.d. and were compared using the t test or the Wilcoxon’s rank-sum test. Categorical variables were expressed as frequencies and percentages and were compared using the χ2 test or the Fisher’s exact test. All clinical variables showing statistical difference in univariate analysis were entered into multivariable analysis. Multivariate logistic regression with log rank test was performed for the determination of the baseline predictors for the development of RH, using backward stepwise selection. All statistical analyses were performed using SAS version 9.1 (SAS Institute Inc., Cary, NC, USA).
Characteristics of the study population
Between December 2008 and November 2011, the HIT registry recruited 2557 patients with uncontrolled hypertension from 43 major hospitals across Korea, including rural and urban areas. As shown in the flow diagram of Figure 1, 118 patients who violated the inclusion or exclusion criteria and 332 patients who were not followed up to 6 months were excluded from the analysis. Of the 2107 eligible patients, we identified 404 patients fulfilling the criteria of RH and 837 patients controlled with one or two antihypertensive medications (non-RH). As a result, approximately 20% of the study population was apparently identified as treatment-resistant after 6-month antihypertensive treatment.
Baseline characteristics of the study patients regarding whether or not they have apparent treatment RH are shown in Table 1. In a total of 1241 patients, mean age was 69.7±6.3 years. Less than a half of patients (42.1%) were male, and the average body mass index (BMI) was 24.9±3.3 kg m−2. The majority of patients (85.7%) were aware of their status of high BP at baseline. When we analyzed target organ damage, the prevalence of heart disease, stroke and microalbuminuria in patients with RH was 54.8, 8.6 and 19.3%, respectively.
Medication data about the proportion of antihypertensive medication administered in patients with resistant and non-RH are also shown in Table 1. By definition, diuretics were administered to all the patients with RH. The next widely used antihypertensive medications in Korea were angiotensin receptor blockers, calcium channel blockers and beta blockers, in sequence.
In univariate analysis for comparison with patients with non-RH, patients apparently identified as treatment-resistant hypertension were more likely to be aware of their status of high BP at baseline and likely to have high BMI and SBP. Patients with RH were also more likely to have DM, hyperlipidemia, cerebrovascular accident, microalbuminuria, chronic kidney disease and retinopathy.
By multivariate logistic regression, the independent predictors of RH are shown in Table 2. In univariate analysis between resistant and non-resistant hypertensives, clinical variables showing statistical differences except medications were entered into a multiple logistic regression model, which are described in Table 1. Clinical characteristics associated with the development of RH after 6-month treatment were awareness of hypertension, high baseline SBP (⩾160 mm Hg), microalbuminuria, DM and high BMI (⩾24 kg m−2). Notably, awareness of hypertension was the strongest predictor (odds ratio (OR) 4.47, 95% confidence interval (CI) 2.45–8.15) in the development of RH.
In the subgroup analysis of drug-naive patients who had completed 6-month antihypertensive treatment, of the 310 drug-naive patients, 34 patients were categorized as RH. We compared them with188 patients controlled with one or two antihypertensive medications (non-RH). In drug-naive patients, 10.9% were apparently identified as RH within 6 months from initial antihypertensive treatment. Baseline characteristics of drug-naive patients were similar to those of total patients. However, they appeared to have less target organ damage than patients who had previously been treated with antihypertensive medications. In univariate analysis, drug-naive patients apparently identified as RH had higher baseline SBP and greater awareness of their status of high BP (Table 3). By multivariate analysis, awareness of hypertension was the only independent predictor for the development of RH (OR 5.16, 95% CI 2.27–11.71) in drug-naive patients (Table 4).
Results of the current study showed that among elderly Korean hypertensives who visited secondary hypertension clinics for high BP 19.1% of the study patients were identified as resistant after 6-month antihypertensive treatment. Notably, 10.9% of the drug-naive patients were categorized as resistant within 6 months from initial treatments. Patients with apparent treatment RH were more likely to be aware of their status of high BP before treatment and to have high baseline SBP (⩾160 mm Hg), microalbuminuria, DM and high BMI (⩾24 kg m−2), compared with patients controlled with one or two antihypertensive agents.
According to retrospective large cohort studies of hypertension control in Spain and the United States, approximately 7.6% and 8.9% of the hypertensive population met the criteria of RH.2, 7 In another cohort study that enrolled 205 750 subjects from the incident hypertensive population, 1.9% of patients (n=3960) who were undergoing treatment developed RH during a median 1.5 years from initial treatment.4 In comparison with these large cohort studies, the prevalence and incidence of RH in the current study were higher than expected. These results might be attributed to different study populations, and elderly hypertensive patients enrolled in the current study were known to be associated with high prevalence of RH and were considered difficult to control due to vascular stiffness.11, 12 Also, we enrolled elderly patients who visited secondary hypertension clinics, not in the primary care, for high BP. Such a selective inclusion criteria could contribute to the different prevalence of RH. Some patients could be overdiagnosed or misidentified as RH due to a lack of assessing orthostatic hypotension, which is prevalent in the elderly patients. Although we intended to investigate whether or not orthostatic hypotension is during follow-up visit, most of BP measurements were performed in a sitting position.
Large observational registries and clinical trials have reported an association of clinical characteristics such as old age, high baseline BP, target organ damages, diabetes, obesity and black race with RH.4, 7, 13 Notably, in the current study, awareness of hypertension at baseline was the most powerful independent predictor for the development of RH. In drug-naive patients, awareness of hypertension at baseline was the only predictor for the development of RH. Awareness of hypertension could reflect the chronicity or duration of hypertension. In a large observational study conducted in Spain, longer duration of hypertension showed an association with development of RH.7 This might be interpreted that chronicity of inadequately treated or untreated hypertension could impact the development of RH and following antihypertensive treatments.
In the Anglo-Scandinavian Cardiac Outcome Trial trial, high baseline SBP could stratify the risk of development of RH.13 Results of the current study also showed an association of stage II hypertension (SBP⩾160 mm Hg) at baseline with RH. This could justify the current European guideline that allowed combination therapy to be the first choice as a therapeutic approach in high-risk patients.14
Microalbuminuria, one type of target organ damage, has been suggested to be more prevalent in resistant hypertensives. According to data from the Spanish Ambulatory Blood Pressure Registry, patients with RH had more microalbuminuria when compared with hypertensive patients controlled on three or less drugs (27.7% vs 15.4%, P<0.001).6 In terms of prognostic importance of microalbuminuria, in a prospective study from Brazil, the presence of both reduced estimated glomerular filtration ratio and microalbuminuria significantly increased cardiovascular risk in 531 patients with RH.15 Furthermore, regression of microalbuminuria was associated with a 27% lower risk and development with a 65% higher risk of having a cardiovascular risk in the same study group.16 Although we did not evaluate the cardiovascular mortality due to short-term follow-up, we could verify the predictive power of microalbuminuria in terms of becoming treatment-resistant hypertensives after antihypertensive treatments. Among the types of target organ damages, microalbuminuria might be regarded as one of the most important diagnostic and prognostic marker of RH.
Although use of angiotensin receptor blockers was unexpectedly high in the current study, which may seem to be against the US or European guidelines, this is probably to reflect previous antihypertensive medications when patients were <60 years (Table 1). Ages of more than a half of patients were between 60 and 69 years. Most of patients (82.2%) had been treated with antihypertensive medications before they were enrolled in the current study.
This study evaluated observational data, and treatment strategies such as lifestyle modifications and antihypertensive medications were not controlled. Therefore, medication adjustments and treatment outcomes could be influenced by different treatment strategies. Also, concurrent use of medications such as nonsteroidal anti-inflammatory drugs, alcohol consumption and dietary sodium intake were not controlled and could affect BP control.
Second, some reports have highlighted the performance of ambulatory BP monitoring for identification of true RH; however, ambulatory BP monitoring was not performed in the current study. Although BP was measured by a non-physician using automated BP device in order to minimize the effect of white-coat hypertension, its influence could not be completely excluded. However, target organ damage such as cerebrovascular accident, microalbuminuria, chronic kidney disease and retinopathy in resistant hypertensives was significantly higher, compared with those of non-resistant hypertensives; from this aspect, pseudohypertension might be excluded acceptably.
Third, although it would be preferable to use a quantitative method for determining microalbuminuria corrected by creatinine, we used the MICRAL test as a semiquantitative method detecting microalbuminuria. However, the performance of MICRAL test as a screening for microalbuminuria was validated in previous reports.17, 18, 19 One of the reports showed that in 396 urine specimens for 132 patients sensitivity was 91% and specificity 96% for a discriminating albumin level of 20 mg l−1. Correlation with quantitative values was reasonable (r=0.73).18
Finally, we excluded 422 patients controlled with three drugs and 369 patients uncontrolled with one or two medications in the analysis. This could be considered as large data loss and might affect the study results. The reason why we did exclude 422 patients controlled with three drugs is that we considered relatively short ‘6-month’ treatment period in the current study. If the antihypertensive treatments were extended to 1 or 2 years, some of this group might be classified as RH. This could also obscure the analytical results. With regard to 369 patients uncontrolled with one or two medications, this reflects truly therapeutic inertia to the elderly patients. One of the previous reports showed that old age (>65 years) was the one of the independent factors associated with therapeutic inertia.20
Findings of this study demonstrated that among elderly Korean hypertensives who visited secondary hypertension clinics for high BP the prevalence of RH was approximately 20%. Clinical characteristics such as awareness of hypertension, high baseline SBP (⩾160 mm Hg), microalbuminuria, DM and high BMI (⩾24 kg m−2) were shown an association with identification of RH after 6-month antihypertensive treatments. In drug-naive patients, awareness of hypertension implying chronicity of hypertension possibly was the only independent predictor of RH. These findings could have an impact on the evaluation and antihypertensive treatment of elderly hypertensive patients.
Calhoun DA, Jones D, Textor S, Goff DC, Murphy TP, Toto RD et al. Resistant hypertension: diagnosis, evaluation, and treatment: a scientific statement from the American Heart Association Professional Education Committee of the Council for High Blood Pressure Research. Circulation 2008; 117 (25): e510–e526.
Persell SD . Prevalence of resistant hypertension in the United States, 2003–2008. Hypertension 2011; 57 (6): 1076–1080.
Egan BM, Zhao Y, Axon RN, Brzezinski WA, Ferdinand KC . Uncontrolled and apparent treatment resistant hypertension in the United States, 1988 to 2008. Circulation 2011; 124 (9): 1046–1058.
Daugherty SL, Powers JD, Magid DJ, Tavel HM, Masoudi FA, Margolis KL et al. Incidence and prognosis of resistant hypertension in hypertensive patients. Circulation 2012; 125 (13): 1635–1642.
de Souza F, Muxfeldt ES, Salles GF . Prognostic factors in resistant hypertension: implications for cardiovascular risk stratification and therapeutic management. Expert Rev Cardiovasc Ther 2012; 10 (6): 735–745.
de la Sierra A, Banegas JR, Oliveras A, Gorostidi M, Segura J, de la Cruz JJ et al. Clinical differences between resistant hypertensives and patients treated and controlled with three or less drugs. J Hypertens 2012; 30 (6): 1211–1216.
de la Sierra A, Segura J, Banegas JR, Gorostidi M, de la Cruz JJ, Armario P et al. Clinical features of 8295 patients with resistant hypertension classified on the basis of ambulatory blood pressure monitoring. Hypertension 2011; 57 (5): 898–902.
Anwar YA, Giacco S, McCabe EJ, Tendler BE, White WB . Evaluation of the efficacy of the Omron HEM-737 IntelliSense device for use on adults according to the recommendations of the Association for the Advancement of Medical Instrumentation. Blood Press Monit 1998; 3 (4): 261–265.
Pickering TG, Hall JE, Appel LJ, Falkner BE, Graves J, Hill MN et al. Recommendations for blood pressure measurement in humans and experimental animals: part 1: blood pressure measurement in humans: a statement for professionals from the Subcommittee of Professional and Public Education of the American Heart Association Council on High Blood Pressure Research. Circulation 2005; 111 (5): 697–716.
Levey AS, Coresh J, Greene T, Stevens LA, Zhang YL, Hendriksen S et al. Using standardized serum creatinine values in the modification of diet in renal disease study equation for estimating glomerular filtration rate. Ann Intern Med 2006; 145 (4): 247–254.
Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL et al. Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension 2003; 42 (6): 1206–1252.
Sarafidis PA, Bakris GL . State of hypertension management in the United States: confluence of risk factors and the prevalence of resistant hypertension. J Clin Hypertens (Greenwich) 2008; 10 (2): 130–139.
Gupta AK, Nasothimiou EG, Chang CL, Sever PS, Dahlof B, Poulter NR . Baseline predictors of resistant hypertension in the Anglo-Scandinavian Cardiac Outcome Trial (ASCOT): a risk score to identify those at high-risk. J Hypertens 2011; 29 (10): 2004–2013.
Mancia G, De Backer G, Dominiczak A, Cifkova R, Fagard R, Germano G et al. 2007 guidelines for the management of arterial hypertension: the Task Force for the Management of Arterial Hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). Eur Heart J 2007; 28 (12): 1462–1536.
Salles GF, Cardoso CR, Pereira VS, Fiszman R, Muxfeldt ES . Prognostic significance of a reduced glomerular filtration rate and interaction with microalbuminuria in resistant hypertension: a cohort study. J Hypertens 2011; 29 (10): 2014–2023.
Salles GF, Cardoso CR, Fiszman R, Muxfeldt ES . Prognostic importance of baseline and serial changes in microalbuminuria in patients with resistant hypertension. Atherosclerosis 2011; 216 (1): 199–204.
Jury DR, Mikkelsen DJ, Glen D, Dunn PJ . Assessment of Micral-Test microalbuminuria test strip in the laboratory and in diabetic outpatients. Ann Clin Biochem 1992; 29 (Pt 1): 96–100.
Spooren PF, Lekkerkerker JF, Vermes I . Micral-Test: a qualitative dipstick test for micro-albuminuria. Diabetes Res Clin Pract 1992; 18 (2): 83–87.
Bangstad HJ, Try K, Dahl-Jorgensen K, Hanssen KF . New semiquantitative dipstick test for microalbuminuria. Diabetes Care 1991; 14 (11): 1094–1097.
Redon J, Coca A, Lazaro P, Aguilar MD, Cabanas M, Gil N et al. Factors associated with therapeutic inertia in hypertension: validation of a predictive model. J Hypertens 2010; 28 (8): 1770–1777.
The HIT study was supported by Sanofi, Korea by providing web pages/server and collecting patients’ data.
The authors declare no conflict of interest.