A Study of Losartan, Alone Or with Hydrochlorothiazide Vs Nifedipine GITS in Elderly Patients with Diastolic Hypertension

Home , Moexipril

Journal of Human Hypertension (1998) 12, 693–699  1998 Stockton Press. All rights reserved 0950-9240/98 $12.00 http://www.stockton-press.co.uk/jhh ORIGINAL ARTICLE A study of losartan, alone or with hydrochlorothiazide vs nifedipine GITS in elderly patients with diastolic hypertension

PR Conlin1, M Elkins2, C Liss3, AJ Vrecenak2, E Barr2 and JM Edelman2 1Endocrinology-Hypertension Division, Brigham and Women’s Hospital, Medical Service, Brockton/West Roxbury VA Medical Center and Harvard Medical School, Boston, MA; 2Clinical Development Department, US Human Health; 3BARDS, Merck Research Labs, Merck and Co, Inc, West Point, PA, USA

We conducted a randomised, double-blind, parallel had DBP reductions of −14, −15, and −15 mm Hg, design study comparing the efficacy and tolerability of respectively. There were no significant differences in the angiotensin II receptor antagonist, losartan, alone or the DBP response between the treatment groups except with low-dose hydrochlorothiazide (HCTZ) to the dihyd- at week 4 (P Ͻ 0.05). Similar reductions in systolic BP ropyridine calcium channel blocker, nifedipine GITS (SBP) between the two treatment groups were observed (gastro-intestinal therapeutic system), in elderly at all time points. The percentages of patients in the two patients (у65 years old) with a diastolic blood pressure treatment groups reaching goal DBP (Ͻ90 mm Hg or (DBP) between 95 and 115 mm Hg. After a placebo wash DBP у90 mm Hg with a reduction from a baseline of у10 out period, 140 patients were randomly assigned to mm Hg) were comparable (81% on the losartan regimen receive either losartan 50 mg or nifedipine GITS 30 mg. and 90% on the nifedipine GITS regimen). There were Patients were evaluated at 4-week intervals during a 12- significantly more adverse events reported in patients week treatment period. Patients receiving losartan had receiving nifedipine GITS when compared to the losar- HCTZ 12.5 mg added and increased to 25 mg to reduce tan regimen (54% vs 36%, P Ͻ 0.05). A patient-reported DBP Ͻ90 mm Hg. Patients receiving nifedipine GITS had symptom inventory also showed that swollen ankles their dose increased to 60 mg and 90 mg to reduce DBP was bothersome in significantly more patients treated Ͻ90 mm Hg. Efficacy, tolerability and quality of life were with the nifedipine GITS regimen when compared to the assessed during the 12 weeks on each regimen. losartan regimen (24% vs 5%, P = 0.001). Thus, in elderly Patients treated with the losartan regimen (n = 73) had patients with diastolic hypertension, a regimen of losar- reductions in trough sitting DBP of −10, −13, and −13 tan alone or with HCTZ has similar efficacy to a regimen mm Hg after 4, 8, and 12 weeks of therapy, respectively. of nifedipine GITS with greater tolerability and less Patients receiving the nifedipine GITS regimen (n = 67) symptom bother due to swollen ankles.

Keywords: hypertension; aged; losartan; nifedipine; angiotensin II receptor blocker; calcium channel blocker; quality of life

Introduction improvement in the incidence of stroke and heart attack as well as death due to these cardiovascular Treatment of high blood pressure (BP) in patients complications.3–5 with essential hypertension has been clearly shown While more than 50% of individuals Ͼ65 years of to reduce cardiovascular events (ie, myocardial 2 1 age have isolated systolic or diastolic hypertension infarction and stroke). These findings are even more the number of patients currently being treated for applicable to older individuals in whom the preva- hypertension is much smaller.6 There are certainly lence of cardiovascular disease is much greater and, many reasons older patients are not adequately therefore, the benefits of treatment are potentially 2 treated, including access to care, cost of medications more substantial. Several large clinical trials of and drug-related side effects.7–9 Evidence also sug- anti-hypertensive therapy in elderly individuals gests that medication intolerability can contribute to with isolated systolic or diastolic hypertension have drug discontinuations.10 The angiotensin II receptor demonstrated this point. BP reduction (systolic blocker, losartan, is indicated for the treatment of and/or diastolic) is associated with significant essential hypertension. By blocking the interaction

of angiotensin II with its receptor (AT1 subtype), losartan reduces BP in a smooth and sustained fashion. Correspondence: Dr Paul R Conlin, Endocrinology-Hypertension In controlled clinical studies, the overall incidence Division, Brigham and Women’s Hospital, 221 Longwood 11 Avenue, Boston, MA 02115, USA of side effects has been comparable to placebo. As Received 11 November 1997; revised 19 May 1998; accepted 25 with other anti-hypertensive agents, losartan can be May 1998 combined with hydrochlorothiazide (HCTZ) to Treatment of hypertension in elderly patients PR Conlin et al 694 further decrease BP when therapeutic goals have not Patients whose diastolic BP (DBP) measurements been achieved with monotherapy. The efficacy of were between 95 and 115 mm Hg and did not differ the two agents combined has been shown to be addi- by more than 7 mm Hg at both their 2- and 4-week tive.12 placebo period visits were deemed eligible for entry The purpose of this study was to compare the into the active treatment period. Each patient was safety, efficacy and tolerability of losartan alone or randomised to receive either losartan or nifedipine with HCTZ vs a regimen of nifedipine GITS (Gastro- GITS in double-blind fashion. Evaluations were per- Intestinal Therapeutic System) in the treatment of formed every 4 weeks and included measurements elderly patients (у65 years old) with diastolic of vital signs, safety monitoring, and quality of life hypertension. Nifedipine GITS is a preparation of assessment. All observed or volunteered adverse nifedipine which is formulated as a controlled- experiences were recorded and a determination was release tablet for once daily dosing. This agent was made by the investigator whether they were drug- chosen as a comparator since recent trends in medi- related. At the conclusion of the active treatment cation choices for elderly individuals with essential period a follow-up clinical exam, laboratory safety hypertension have shown a substantial rise in the assessment and quality of life questionnaire were use of calcium channel blockers.13 Study objectives completed. were to determine the ability of these regimens to The goal of the study was to have patients achieve achieve target BP (DBP Ͻ90 mm Hg) and the overall a sitting DBP Ͻ90 mm Hg with active treatment. tolerability of each treatment scheme, as measured Therefore, during follow-up if the patients’ DBP was by investigator reported adverse events and patient у90 mm Hg their dose of study medication was reported quality of life. titrated as follows: losartan 50 mg daily → losartan 50 mg/HCTZ 12.5 mg daily → losartan 50 mg/HCTZ → Subjects and methods 25 mg daily or nifedipine GITS 30 mg daily 60 mg daily → 90 mg daily. The first titration step usually Participants in this study (n = 140) were men and occurred at week 4 of active treatment although women aged у65 years with known diastolic hyper- early titration was allowed for patients with DBP tension. Twenty-eight clinical centres participated Ͼ110 mm Hg after at least 2 weeks of active treat- in the study (see Appendix). The protocol was ment. reviewed and approved by Institutional Review All medication and matching placebo were taken Boards for Human Studies at each of the participat- at the same time of day, usually between 8.00 and ing sites and written, informed consent was 10.00 am. To maintain the blinding of patients and obtained from each participant prior to enrolment. investigators as to study drug assignments, medi- Initial screening involved a medical history, physi- cation was dispensed as active drug and matching cal examination and basic laboratory studies. Sub- placebo. Each pill was taken once daily. For jects were excluded if they had a history of signifi- example, patients randomised to the losartan regi- cant cardiovascular, cerebrovascular, renal or men initially received losartan 50 mg tablets along hepatic disease, or secondary hypertension. with placebo matching nifedipine GITS 30 mg. If dose titration was required, they were provided a single pill containing losartan 50 mg/HCTZ 12.5 mg Study design and placebo matching nifedipine GITS 60 mg. At 8 This was a randomised, double-blind, parallel weeks of treatment, if further dose titration was design multicentre clinical trial comparing the anti- required the patients were provided losartan 50 mg hypertensive efficacy, tolerability and effects on tablets and HCTZ 25 mg tablets and placebo match- quality of life during therapy with losartan 50 mg ing nifedipine GITS 90 mg. Those patients random- alone or with low dose HCTZ (12.5 or 25 mg) vs ised to the nifedipine GITS regimen received active nifedipine GITS 30, 60, or 90 mg in elderly patients drug along with matching placebo for losartan, with essential hypertension. All patients were with- losartan/HCTZ and HCTZ, respectively. drawn from prior anti-hypertensive medications and entered into a 4-week baseline period during which Quality of life assessment they received one tablet each of placebo matching losartan and nifedipine GITS daily. The patients A battery of scales questionnaire was used to evalu- were monitored every 2 weeks for vital signs and ate differences in health related quality of life. Six adverse experiences. Laboratory assessment was domains were included: overall health perceptions, obtained after 4 weeks on a placebo. Quality of life psychological well being, social functioning, sleep questionnaires were administered at each study disturbance, cognitive functioning, and sexual func- visit. tioning. A symptom inventory was also used to All clinic personnel responsible for obtaining BP assess disease and treatment related complaints. measurements were certified through a standardised Symptom bother was assessed at each visit during training programme using the American Heart the placebo and active treatment periods. The Association guidelines for BP measurement. BP was patients were asked to report the amount of bother measured in the sitting position after at least 5-min they experienced from 32 different symptoms by rat- of rest. All measurements were taken 22–26 h after ing them as ‘not at all’, ‘little’, ‘moderately’, ‘quite a the last dose of study medication. Three readings bit’, or ‘extremely’. Symptom bother was considered were taken at 1-min intervals and averaged to pro- absent if it was reported as ‘not at all’ or ‘little’ dur- vide a mean value. ing the baseline period and present if symptom Treatment of hypertension in elderly patients PR Conlin et al 695 bother was considered as ‘moderately’, ‘quite a bit’ Table 1 Demographic and clinical characteristics of study sub- or ‘extremely’ at any time during the study. A jects change was defined as the presence of bother from Losartan Nifedipine a particular symptom which was absent at baseline regimen GITS but present at any time during the study. For data (n = 73) regimen analysis symptom bother was recorded as present at (n = 67) its first occurrence and was not dependent upon its continuation throughout the study. Age (years) Male 71 ± 471± 4 Female 72 ± 572± 5 Statistical analysis Sex Male 47 (64%) 39 (58%) The study was designed to have 90% power to Female 26 (36%) 28 (42%) detect a 5 mm Hg difference between treatment Race groups in the change from baseline DBP as signifi- African American 19 (26%) 16 (24%) cant at ␣ Ͻ0.05. All data analysed and presented are White 51 (70%) 45 (67%) based upon all treated patients (intent to treat Hispanic 2 (3%) 5 (7%) analysis). For patients who withdrew prior to com- Asian 1 (1%) 1 (1%) pletion of the study, the last value recorded on treat- Baseline blood pressure (mm Hg) Systolic 164 ± 18 164 ± 17 ment was ‘carried forward’ to each subsequent ± ± observation period for inclusion in the analysis. Diastolic 100 5 100 5 Baseline comparability of the two treatment groups Mean ± s.d. with respect to demographic and clinical characteristics was assessed using Student’s t-test and the Fisher Exact Test. Changes in BP were analysed between the two treatment groups at baseline. Of the employing the paired t-test for within group differ- 140 patients who entered the treatment phase, 88% ences and analysis of variance (ANOVA) for of losartan-treated patients (n = 64) and 90% of = between group differences. Summary statistics were nifedipine GITS-treated patients (n 60) completed calculated and comparisons made between the end the study. of baseline and each period of observation for the Each of the treatment regimens produced a sig- two treatment regimens (ie, weeks 4, 8, and 12). Dif- nificant fall in both systolic and diastolic BP at week ferences within and between treatment groups were 12. These results are shown in Table 2. The DBP considered significant at ␣ Ͻ0.05. response at 4 weeks of therapy showed a signifi- The health related quality of life questionnaire cantly greater effect with nifedipine GITS 30 mg − ± − ± was designed to have 95% power to detect a change than with losartan 50 mg ( 14 9 vs 10 9 mm Hg, of 7–13% (depending on the domain) as significant P Ͻ 0.05; Figure 1). Subsequent DBP measurements at ␣ Ͻ0.05. Differences between treatment groups for at weeks 8 and 12 of double-blind treatment did not the health related quality of life scales were com- differ. This was in part related to the addition of pared using an ANOVA model and within group HCTZ to the losartan regimen or dose titration of changes were analysed using a paired t-test. The nifedipine GITS for patients whose DBP did not fall domains of general health and sexual functioning below 90 mm Hg after 4 weeks of treatment. At the were emphasised. A Bonferroni correction for multi- end of the 12-week active treatment period 32 plicity was employed for comparing treatments with patients (44%) remained on losartan alone and 41 regard to these two domains and a P Ͻ 0.025 was patients (56%) required addition of HCTZ. Twenty- considered significant. Regarding symptom bother, six patients (36%) were titrated to losartan between group differences were compared using the 50 mg/HCTZ 12.5 mg and 15 patients (20%) were chi-square test and within group changes were ana- titrated to losartan 50 mg/HCTZ 25 mg. In the nifedi- lysed with the McNemar’s test. For all symptoms, pine GITS-treated patients, 44 (65%) remained on treatment differences were considered significant at 30 mg daily, 16 patients (24%) were titrated to ␣ Ͻ0.05 except for the symptoms of swollen ankles, 60 mg and seven patients (11%) were titrated to headache, and flushing, which were prospectively 90 mg daily. The number of patients requiring dose considered to be more likely in nifedipine-treated titration during the study (losartan regimen 41/73; patients. In this case a Bonferroni correction for nifedipine GITS regimen 23/67) was significantly Ͻ multiplicity was employed and a P Ͻ 0.017 was con- different (P 0.01). All patients had a significant sidered significant. fall in DBP during the study regardless of final study medication(s) assignment, as shown in Figure 2. Those patients who required the last dose titration Results step tended to have the highest baseline DBP and Efficacy of treatment the smallest decrement in DBP. Goal BP was prospectively defined as a DBP Ͻ90 A total of 140 patients were entered into the trial; mm Hg or DBP у90 mm Hg with a decrease of у10 73 patients were randomised to the losartan regimen mm Hg during treatment. The number and percent- and 67 patients to the nifedipine GITS regimen. age of patients achieving goal BP were similar Table 1 provides a summary of the demographic and between the two treatment groups. Fifty patients clinical characteristics of the patients in each treat- (68%) treated with the losartan regimen achieved a ment group. No significant differences were found DBP Ͻ90 mm Hg; an additional nine patients (12%) Treatment of hypertension in elderly patients PR Conlin et al 696 Table 2 Systolic and diastolic blood pressure response to treatment

Treatment No. Week Systolic Change from Diastolic Change from (mm Hg) baseline (mm Hg) baseline

Losartan regimen 73 Baseline 164 ± 18 100 ± 5 73 12 149 ± 15* −16 ± 18 87 ± 8* −13 ± 8 Nifedipine GITS regimen 67 Baseline 164 ± 17 100 ± 5 67 12 145 ± 17* −19 ± 16 85 ± 7* −15 ± 7

Mean ± s.d. *P Ͻ 0.05 vs baseline.

Figure 1 Mean changes in trough sitting DBP by weeks of treatment. All changes from baseline with each of the regimens were signiﬁcantly different (P Ͻ 0.001). *P Ͻ 0.05 vs losartan regimen.

had DBP у90 mm Hg with a decrement of у10 mm Hg. In the nifedipine GITS-treated patients 54 (82%) achieved a DBP Ͻ90 mm Hg with five additional patients (8%) having DBP у90 mm Hg with a decrement of у10 mm Hg. Overall, 59 patients (81%) in the losartan regimen and 59 patients (90%) in the nifedipine GITS regimen achieved goal DBP (P = NS). Figure 2 Mean changes in trough sitting DBP during the 12 weeks of study based on final treatment received. All changes from base- Analysis of BP responses in whites and African line to the end of study were significantly different. Tests of sig- Americans was also conducted. Of the 140 parti- nificance were not applied to compare the BP responses between cipants in the study 96 were white and 35 were the different treatment subgroups. Assignment by final treatment African American. The randomised treatment received was not randomised but based on an individual’s BP response. Therefore, observed differences may not be attributed assignments resulted in 19 African American to treatment effects alone. *P Ͻ 0.001 vs baseline. Los = Losartan, patients receiving losartan and 16 receiving nifedip- HCTZ = hydrochlorothiazide, Nif = nifedipine GITS. ine GITS. This analysis was not pre-specified in the protocol and the sample sizes were small. Nonethe- less, the decrease in DBP at week 12 was similar with nifedipine did not show a dose dependency. between the two racial groups (Table 3). Of the seven patients in whom oedema was reported five patients were receiving 30 mg daily, one patient was taking 60 mg daily and one patient was taking Safety and tolerability of treatment 90 mg daily. Those adverse events which were con- All reported adverse events (which included those sidered drug-related by the investigators did not dif- considered by the investigators to be drug-related) fer between the two regimens (losartan 19%, nifedi- were significantly more frequent in patients receiv- pine 21%). No significant laboratory test ing nifedipine GITS than those treated with the los- abnormalities (including serum potassium and uric artan regimen (54% vs 36%, P Ͻ0.05). Oedema acid levels) were noted in patients treated with (losartan 5%, nifedipine 9%), headache (losartan either the losartan regimen or nifedipine GITS. Six- 5%, nifedipine 7%), and diarrhoea (losartan 4%, teen patients discontinued the study prior to its nifedipine 3%) were the most common adverse completion. Early withdrawal was attributed to an events reported. There were no significant differ- adverse event in eight of these patients (four each in ences between treatment groups in the reporting of the losartan and nifedipine-treated groups, these specific symptoms. The incidence of oedema respectively). Treatment of hypertension in elderly patients PR Conlin et al 697 Table 3 Blood pressure response by race

Week Race Losartan regimen Nifedipine GITS regimen

No. SBP DBP No. SBP DBP (mm Hg) (mm Hg) (mm Hg) (mm Hg)

Baseline African American 19 164 ± 19 100 ± 5 16 167 ± 17 101 ± 6 12 19 146 ± 12* 85 ± 7* 16 156 ± 20* 87 ± 7* Baseline White 51 165 ± 17 100 ± 5 45 162 ± 17 100 ± 4 12 51 151 ± 15* 89 ± 8* 45 141 ± 15* 85 ± 7*

Mean ± s.d. SBP = systolic blood pressure (mm Hg); DBP = diastolic blood pressure (mm Hg). *P Ͻ 0.05 vs baseline.

The study subjects were asked to complete a was comparable between the two treatment regi- health related quality of life questionnaire at each mens. visit during the study. There were no significant The DBP response to losartan 50 mg alone was changes from baseline to the end of the study in less than with nifedipine GITS 30 mg and this was overall health perception, cognitive functioning, reflected in more patients in the losartan regimen social functioning, sexual functioning and psycho- requiring addition of low-dose HCTZ to achieve goal logical well-being in either treatment group. Nifedi- BP. This observation could be a function of the pro- pine-treated patients reported significant worsening tocol’s titration at 4 weeks of therapy for those (P Ͻ 0.05) from baseline for sleep disturbances at patients with DBP у90 mm Hg or might also be due weeks 4, 8 and 12. The losartan-treated patients to a true difference in the anti-hypertensive effects showed no significant changes from baseline in this of losartan and nifedipine GITS in elderly subjects. parameter. The symptom inventory portion of the In this latter regard, elderly patients treated with fel- questionnaire queried participants for the presence odipine have been shown to have delayed drug of bother due to 32 different symptoms. A change in clearance resulting in higher drug levels and a symptom bother was considered to have occurred if heightened anti-hypertensive response for a given the symptom was absent at baseline but present at dose of drug.14 It is not known whether a similar any time during the study. Patient-reported symp- effect may be present with nifedipine. Despite the tom bother from swollen ankles was significantly more frequent addition of HCTZ to losartan-treated greater with the nifedipine GITS regimen than with patients, this combination was better tolerated than the losartan regimen (24% vs 5%, P = 0.001). All nifedipine GITS based on the incidence of adverse other symptoms were reported with equal frequency events and symptom bother from swollen ankles. between the two groups including those due to Calcium channel blockers such as nifedipine GITS headache, flushing and sweating (Table 4). have been previously shown to be effective in treat- ing elderly hypertensives,15–17 and this efficacy was Discussion confirmed in the present study. Based on these prior observations it has also been suggested that calcium In this study, a regimen of losartan alone or with channel blockers may be preferred anti-hypertensive low-dose HCTZ had similar anti-hypertensive effi- agents for elderly hypertensives. Indeed, the use of cacy and better tolerability with respect to symptom calcium channel blockers as treatment for elderly bother due to swollen ankles when compared to hypertensive patients has expanded markedly since nifedipine GITS given in step-wise dose increments. their introduction.13 Agents which interrupt the The percentage of patients achieving the therapeutic renin-angiotensin system have also been purported goal of a reduction in DBP to Ͻ90 mm Hg and/or a to not have the same BP-lowering effect as calcium DBP у90 mm Hg with a decrement of у10 mm Hg channel blockers in this group of patients.18 This conclusion has been based on the observation that many elderly patients with essential hypertension Table 4 Symptom bother for selected symptoms during 12 weeks have lower plasma renin activity and, therefore, may of therapy not respond to agents which block the renin-angiotensin system. However, the present study confirms Symptom* Losartan Nifedipine 16,19,20 regimen GITS regimen other studies which have shown that interrup- tion of the renin-angiotensin system in combination Swollen ankles 4 (5%) 16 (24%)† with a diuretic is an effective treatment scheme for Headache 6 (8%) 3 (5%) elderly patients with hypertension. Flushing of face 4 (5%) 5 (8%) Given the similar BP responses between the two Sweating 4 (5%) 9 (14%) treatment regimens, an important finding of the Dry cough 2 (3%) 4 (6%) present study is that the losartan regimen was better *Symptoms were absent at baseline and present during the study. tolerated than nifedipine GITS, as indicated by the Only some of the symptoms that were queried are shown. lower overall incidence of adverse events and less †P Ͻ 0.05 vs losartan regimen. symptom bother due to oedema. Nifedipine GITS Treatment of hypertension in elderly patients PR Conlin et al 698 treated patients experienced more adverse events, a 5 Medical Research Council Working Party. MRC trial of significantly greater incidence of sleep disturbance, treatment of hypertension in older adults: principal and more symptom bother due to swollen ankles results. Br Med J 1992; 304: 405–412. than losartan-treated patients. Other categories 6 Burt VL et al. Prevalence of hypertension in the US assessed by the health related quality of life ques- adult population: Results from the Third National Health and Nutrition Examination Survey, 1988–1991. tionnaires did not differ between the treatment Hypertension 1995; 25: 305–313. groups. It is noteworthy that oedema was reported 7 Finnerty FA, Mattie EC. Hypertension in the inner as an adverse event by the investigators only one- city. I. Analysis of clinic dropouts. Circulation 1973; third as often as patients self-reported symptom 47: 73–75. bother due to swollen ankles. This observation 8 Brand F, Smith R, Brand P. Effect of economic barriers underscores the fact that adverse events are often to medical care on patients’ noncompliance. Public underreported and emphasises the importance of Health Rep 1977; 92: 72–78. good physician-patient communication with regard 9 Shulman NB et al. Financial cost as an obstacle to to drug-related adverse events. hypertension therapy. Am J Med 1986; 81 (Suppl 6C): The recent report of the Joint National Committee 20–24. 10 Curb JD et al. Long-term surveillance for adverse on the Prevention, Detection, Evaluation, and Treat- events of antihypertensive drugs. JAMA 1985; 253: ment of High Blood Pressure (JNC VI) underscores 3263. the high prevalence of hypertension in older per- 11 Goldberg AI, Dunlay MC, Sweet CS. Safety and toler- sons and the importance of its treatment.21 The ability of losartan potassium, an angiotensin II recep- Committee considers optimal drug formulations as tor antagonist, compared with hydrochlorothiazide, those which provide 24 h efficacy with once-daily atenolol, felodipine ER, and angiotensin converting dosing, qualities which are shared by both losartan enzyme inhibitors for the treatment of systemic hyper- and nifedipine GITS. The report notes that angioten- tension. Am J Cariol 1995; 75: 793–795. sin II receptor blockers have similar haemodynamic 12 MacKay JH et al. Losartan and low-dose hydrochloro- effects as ACE inhibitors, with fewer side effects. It thiazide in patients with essential hypertension. A double-blind, placebo-controlled trial of concomitant reaffirms the beneficial effects of thiazide diuretics administration compared with individual compo- in the elderly because of their favourable effects on nents. Arch Intern Med 1996; 156: 278–285. morbidity and mortality. However, we still lack data 13 Monane M et al. Trends in medication choices for on renal and cardiac protection effects of angioten- hypertension in the elderly. The decline of the thiaz- sin II receptor blockers which have been clearly ides. Hypertension 1995; 25: 1045–1051. shown with ACE inhibitors. 14 Wade JR, Sambol NC. Felodipine population dose- In this study the treatment regimen of losartan response and concentration-response relationships in with low-dose HCTZ was equally effective as a regi- patients with essential hypertension. Clin Pharmacol men of nifedipine GITS in elderly patients with Ther 1995; 57: 569–581. elevated DBP. The losartan regimen was better toler- 15 Messerli FH, Grodzicki T. Hypertension and hypertensive emergencies. In: Messerli FH (ed). Cardiovascular ated, as evidenced by a lower overall incidence of disease in the elderly. 3rd edn., Kluwer Academic: adverse events and less symptom bother due to Norwell, MA, 1993, pp 121–161. swollen ankles. Therefore, the use of losartan alone 16 Materson BJ et al. Single-drug therapy for hyperten- or with HCTZ is an effective, well-tolerated treat- sion in men. N Engl J Med 1993; 328: 914–921. ment regimen for elderly patients with essential 17 Neaton JD et al. Treatment of mild hypertension study. hypertension in whom diastolic BP is not reduced JAMA 1993; 270: 713–724. by lifestyle modifications alone. 18 Buhler FR et al. Renin profiling to select antihypertensive baseline drugs: Renin inhibitors for high-renin and calcium entry blockers for low-renin patients. Am Acknowledgements J Med 1984; 77: 36–42. This study was conducted with a grant from Merck 19 Hart W. Lisinopril-hydrochlorothiazide combination and Co, Inc, US Human Health Division. Data man- compared with the monocomponents in elderly hyper- agement and statistical analyses were conducted tensive patients. J Hum Hypertens 1991; 5 (Suppl 2): 85–89. with the assistance of Laurie Gallagher. 20 White WB, Stimpel M. Long-term safety and efficacy of moexipril alone and in combination with hydroch- References lorothiazide in elderly patients with hypertension. J Hum Hypertens 1995; 9: 879–884. 1 Collins R et al. Blood pressure, stroke, and coronary 21 The Sixth Report of the Joint National Committee on heart disease. Part 2. Short term reductions in blood Prevention, Detection, Evaluation, and Treatment of pressure: overview of randomised drug trials in their High Blood Pressure. Arch Int Med 1997; 157: 2413– epidemiological context. Lancet 1990; 355: 827–838. 2446. 2 National High Blood Pressure Education Program Working Group. National High Blood Pressure Edu- cation Program Working Group Report on Hyperten- Appendix sion in the Elderly. Hypertension 1994; 23: 275–285. 3 SHEP Cooperative Research Group. Prevention of The following investigators participated in this stroke by antihypertensive drug treatment in older per- multicentre study: sons with isolated systolic hypertension. JAMA 1991; 265: 3255–3264. 4 Amery A et al. Mortality and morbidity results from Albany, NY James Higgins; Allen Park, MI Martin the European Working Party on Hypertension in the Bermann; Baltimore, MD Matthew Weir; Elderly trial. Lancet 1985; 1: 1349–1354. Birmingham, MI Charles Lucas; Boston, MA Paul Treatment of hypertension in elderly patients PR Conlin et al 699 Conlin, Andrew King; Charleston, SC Walter Brzez- Richard Preston; Minneapolis, MN Richard Grimm; inshi; Chicago, IL Williams Schlueter; Cincinnati, Mobile, AL Marcia Littles; Philadelphia, PA Allen OH Max Reif; Cuyahoga Falls, OH Ross Black; Marks; Phoenix, AZ Paul Rousseau; Portland, OR Fresno, CA James McCarty; Hollywood, FL Harry William Spisak; San Diego, CA Larry Favrot; Seattle, Serfer; Houston, TX S. Noor Rahman; Jenkintown, WA Robert Davidson; Tacoma, WA David Munoz; PA Antoinette Mangione; Little Rock, Ar Vivian Fon- Tampa, FL Hector Fontanet; Vahalla, NY Leonard seca; Manchester, NH Edward Gillie; Miami, FL Medds; Whittier, CA Robert Fiddes.