Evaluation of Indapamide 1.25 Mg Once Daily in Elderly Patients with Mild to Moderate Hypertension

Evaluation of indapamide 1.25 mg once daily in elderly patients with mild to moderate hypertension

R Fiddes1, J Blumenthal2, JE Dawson3, E Dyckman4, PG St John Hammond5, S Harris6, KC Lasseter7, BS Levine8, R Montoro9, AL Niederman10, PH Ratner11, S Rosenblatt12,PD Toth13, J Vergis14 and J Codispoti15 1Private Practice, Whittier, CA; 2Private Practice, Marietta, GA; 3Private Practice, Atlanta, GA; 4Private Practice, Atlanta GA; 5Loma Linda VA Medical Center, Loma Linda, CA; 6South Florida Bioavailability Center, Miami, FL; 7Private Practice, Miami, FL; 8VA Medical Center, Los Angeles, CA; 9Clinical Therapeutics Corp, Coral Gables, FL; 10Private Practice, Fort Lauderdale, FL; 11Sylvana Research, San Antonio, TX; 12Irvine Clinical Research Center, Irvine, CA; 13Midwest Institute for Clinical Research, Inc., Indianapolis, IN; 14Private Practice, Levittown, PA; and 15Rhoˆ ne-Poulenc Rorer Physician, Collegeville, PA, USA

The objective of this study was to evaluate the safety double-blind baseline to the end of 8 weeks of therapy. and efficacy of indapamide 1.25 mg once daily as mono- By week 8 of the double-blind treatment period, indapa- therapy in elderly patients (65 years and older) with mild mide 1.25 mg produced a statistically significant to moderate essential hypertension. Two hundred and (P = 0.0037) decrease in supine DBP of 8.2 mm Hg com- seventy-nine (279) elderly patients were enrolled in a pared to a decrease of 5.3 mm Hg produced in the pla- washout period, during which patients received single- cebo group. Additionally, indapamide 1.25 mg was stat- blind placebo for 4 weeks. Patients demonstrating istically (P = 0.0028) more effective than placebo in supine diastolic pressures between 95 mm Hg and 114 reducing supine systolic BP (SBP) (−10.1 vs −4.2 mm Hg at the end of the 4-week placebo washout period mm Hg). The incidence of drug-related adverse events were entered into the 8-week double-blind treatment during the double-blind treatment period was similar period. Two hundred and four (204) patients qualified between the two treatment groups. for the study and were randomized to the double-blind A low dose of indapamide, 1.25 mg, given once daily treatment; 103 patients received indapamide 1.25 mg for 8 weeks was effective as monotherapy with respect and 101 patients received placebo for 8 weeks. Overall, to BP reduction in an elderly population with mild to 177 patients (92 indapamide and 85 placebo) completed moderate hypertension. Indapamide 1.25 mg was safe the study. The primary efficacy criterion was the mean and generally well tolerated in this elderly patient popu- change in supine diastolic blood pressure (DBP) from lation.

Keywords: indapamide; low-dose; indolines; elderly; monotherapy

Introduction 1.25 mg, compared to placebo in elderly patients with mild to moderate hypertension. Indapamide (Lozol, Rhoˆ ne-Poulenc Rorer Pharma- ceuticals, Collegeville, PA, USA) is an indoline, thiazide-like diuretic with a favourable benefit-to-risk Patients and methods ratio in low doses (2.5 mg and 1.25 mg). Patients Initiation of antihypertensive drug therapy at the lowest possible dose is consistent with optimizing Two hundred and four patients were randomized to safety as well as providing efficacy, especially in the double-blind treatment period, 103 to the inda- elderly patients who may have altered physiologic pamide 1.25 mg group and 101 to the placebo group. functions or concomitant disease processes. One hundred and seventy-seven patients completed Although diuretics are generally well tolerated by the study, 92 (89%) in the indapamide 1.25 mg elderly patients, there exists the potential for hypo- group and 85 (84%) in the placebo group. The demo- tensive, arrhythmic, or generalized adverse systemic graphic characteristics of all randomized patients responses to occur. In keeping with these concepts, are shown in Table 1. this study was undertaken to evaluate the antihypertensive efficacy and safety of low-dose indapamide, Inclusion criteria Patients who were 65 years of age or older with mild to moderate essential hypertension (defined as Correspondence: Robert Fiddes, 12291 E Washington Blvd, Whit- untreated supine diastolic blood pressure (DBP) tier, CA 90606, USA between 95 and 114 mm Hg, inclusive) were Evaluation of indapamide in elderly patients R Fiddes et al 240 Table 1 Patient characteristics single-blind placebo washout period during which a diagnosis of mild to moderate uncomplicated Characteristic Indapamide 1.25 mg Placebo = = hypertension was confirmed (supine DBP between (n 103) (n 101) 95 and 114 mm Hg). Eligible patients were then entered into an 8-week double-blind treatment per- Number and per cent (%) of patients Age (yr) iod where they were randomized to receive indapa- mean 69.4 69.7 mide 1.25 mg or placebo: BP (supine and standing range 65–80 65–83 systolic and diastolic), heart rate and body weight Weight (kg) were measured at each single-blind visit. Three mean 79.5 76.7 supine BP readings were obtained at each visit. range 52–123 46–121 One BP reading was taken immediately after Sex standing, followed by a second BP reading after male 54 (52) 59 (58) 2 min of standing. All BP readings were obtained in female 49 (48) 42 (42) the non-dominant arm using the same BP measuring Race device. At weeks −4 and −1 of the single-blind per- white 58 (56) 59 (58) black 3 (3) 4 (4) iod, fasting clinical laboratory values were obtained hispanic 39 (38) 37 (37) for glucose, creatinine, blood urea nitrogen, haema- oriental 1 (1) 1 (1) tology parameters, and urinalysis. In addition, other 2 (2) 0 patients had a complete physical examination, a 12- Mean baseline 98.8 99.8 lead ECG, and chest X-ray at week −4. At weeks −3, supine diastolic −2, and −1, concomitant medications and any blood pressure adverse events that occurred were recorded. (mm Hg) During the double-blind treatment period, patients were seen every 2 weeks at which time BP, heart rate, and body weight were measured. At screened for enrollment into this study. Other weeks 0 and 4 fasting clinical laboratory parameters, inclusion criteria were: biochemistry, haematology including serum chemistry, haematology, urinal- and urinalysis test results without clinically signifi- ysis, lipids, and electrolytes (week 4 only) were cant abnormalities; serum potassium levels within measured. A 12-lead ECG was repeated at week 0. the range of 3.4 to 5.4 mEq/L without potassium At the final visit (week 8), patients had a complete supplementation; and renal function parameters physical examination, a 12-lead ECG, and fasting within stable levels (BUN р24 mg/dL and creatinine clinical laboratory tests (haematology, biochemistry, р2.0 mg/dL). Patients were to have been previously lipids, and urinalysis). Adverse events and the use on monotherapy treatment for hypertension or to of concomitant medication were recorded through- have been newly diagnosed (no previous drug out the double-blind period. Generally, the investi- therapy) with hypertension. gator gave the patient potassium supplementation Following an explanation of the study and related when the serum potassium level dropped below procedures and prior to the screening visit, all 3.4 mEq/L; however, this was based on the investi- patients were to have signed an Informed Consent gator’s judgment. Form. Statistical analysis Exclusion criteria The primary and secondary efficacy analyses were Patients with malignant or severe hypertension, a carried out on the all-treated patient populations. disease state causing secondary hypertension, or The primary efficacy variable was the mean change patients previously treated with multiple drug ther- in supine DBP from double-blind baseline (week 0) apy for hypertension were excluded from the study. to the end of 8 weeks of therapy. Secondary efficacy Also excluded from the study were patients with criteria were the percentage of patients having a significant diseases of the cardiac conduction sys- supine DBP р90 mm Hg, or a reduction of at least tem, premature ventricular beats considered clini- 10 mm Hg from double-blind baseline to the end of cally significant or requiring treatment, uncontrolled 8 weeks of therapy and to the last study assessment; congestive heart failure, stroke within 1 year with the mean change from double-blind baseline to last residual paresis or other neurologic impairment, sig- study assessment in supine DBP; and the mean nificant renal or vascular disease, treatment with change from double-blind baseline to the end of 8 insulin or lithium, history of recent alcohol or drug weeks of therapy and to the last study assessment abuse, or any clinically significant deviation from in supine systolic BP (SBP), and standing systolic normal in either the physical exam or in labora- and DBP. tory parameters. The significance of a centre by treatment interaction was investigated with a two-way analysis of variance model with centre, treatment group, and Design treatment by centre interaction effects at the con- This was a multicentre, randomized study con- servative level of 10% for primary efficacy only. sisting of two periods: a single-blind placebo wash- Treatment comparisons were performed using a out period and a double-blind treatment period. two-way analysis of variance model with main Patients who met the entry criteria entered a 4-week effects of center and treatment group in supine and Evaluation of indapamide in elderly patients R Fiddes et al 241 standing diastolic and SBPs to the end of 8 weeks mm Hg was observed in the indapamide 1.25 mg of double-blind therapy and to the last study assess- group compared to a mean decrease of 4.2 mm Hg ment. All efficacy comparisons of the treatment in the placebo group (Table 3). Additionally, inda- groups were two-sided at the 0.05 level of signifi- pamide 1.25 mg was statistically (P р0.0029) more cance. Treatment comparisons using the Cochran– effective than placebo in reducing both standing Mantel–Haenszel test, stratified by centre, were per- diastolic and SBP. The respective mean decreases formed for a percentage of patients with a DBP р90 for standing diastolic and SBP, at week 8 were 7.7 mm Hg at the end of the double-blind treatment mm Hg and 11.7 mm Hg in the indapamide 1.25 mg period. group compared to mean decreases of 4.5 mm Hg All safety analyses were based on the all-treated and 2.4 mm Hg in the placebo group. patient population. Descriptive statistics were used Indapamide 1.25 mg was more effective than pla- to summarize safety variables. No formal statistical cebo in producing a treatment response (decrease of test of hypothesis was carried out. Treatment groups supine DBP р90 mm Hg or a reduction of at least were compared with counts and per cents for the 10 mm Hg from double-blind baseline to the end of incidence of adverse events, hypokalemia (defined 8 weeks of therapy). At week 8, the response rate as a potassium level р3.4 mEq/L), death, premature in the indapamide 1.25 mg group was 55% (48/87) discontinuation from study, concomitant medi- compared to 37% (32/87) in the placebo group; this cations, and changes from double-blind baseline to difference between treatment groups was statisti- the last study assessment relative to the normal cally significant (P = 0.020) (Table 4). When defining ranges for haematology and biochemistry para- treatment response in terms of SBP reduction (р140 meters. mm Hg), indapamide 1.25 mg was again more effec- Treatment groups were compared with descriptive than placebo. At week 8, the indapamide tive statistics for mean changes from double-blind 1.25 mg group response rate was 32% (28/87) com- baseline to the last study assessment for haema- pared to 14% (12/87) in the placebo group; the dif- tology and biochemistry parameters, ECG data, api- ference between treatment groups was statistically cal heart rate and body weight. significant (P = 0.002) (Table 5). Indapamide 1.25 mg was well tolerated in this Results elderly patient population. The incidence of all adverse events during the treatment period was A total of 204 patients were randomized to the slightly lower in the indapamide 1.25 mg group double-blind treatment period, 103 to the indapam- compared to the placebo group (31% vs 38%). The ide 1.25 mg group and 101 to the placebo group. One most frequently reported adverse event in the inda- hundred and seventy-seven patients completed the pamide 1.25 group and the placebo group was head- study, 92 (89%) of the indapamide group and 85 ache (5% vs 7%) (Table 6). The incidence of drug- (84%) of the placebo group. Twenty-seven patients related adverse events was similar between the inda- (11 indapamide, 16 placebo) were prematurely dis- pamide 1.25 mg and placebo groups (9% vs 12%). continued from the study for the following reasons: Of the drug-related adverse events, leg cramps were adverse events (4 indapamide, 6 placebo), ineffec- reported by the highest percentage of patients, 4% tive therapy (1 indapamide, 6 placebo), protocol of the indapamide-treated patients. deviations (1 indapamide, 2 placebo), withdrawn Five patients in the indapamide 1.25 mg treatment consent (1 indapamide, 1 placebo), lost to follow-up group and six in the placebo treatment group were (2 indapamide), and other (2 indapamide, 1 discontinued due to adverse events. Seven patients placebo). reported serious adverse events, two in the indapa- Indapamide 1.25 mg was statistically (P = 0.0037) mide 1.25 mg group and five in the placebo group. more effective than placebo in reducing supine DBP. None of the serious adverse events were considered At week 8, a mean decrease of 8.2 mm Hg was by the investigator to be related to study drug observed in the indapamide 1.25 mg group com- administration. One patient in the indapamide pared to a mean decrease of 5.3 mm Hg in the pla- 1.25 mg treatment group died due to underlying cebo group (Table 2). arteriosclerosis as assessed by the investigator, and Indapamide 1.25 mg was also statistically one patient in the placebo treatment group died of (P = 0.0028) more effective than placebo in reducing myocardial infarction. In both cases, the cause of supine SBP. At week 8, a mean decrease of 10.1 death was not related to study drug administration.

Table 2 Summary of mean changes from baseline (week 0) to week 8 and endpoint in supine diastolic blood pressure all-treated patients

Week Indapamide 1.25 mg Placebo

n Mean value Mean change n Mean value Mean change P-value (mm Hg) from baseline (mm Hg) from baseline (standard (standard deviation) deviation)

Week 0 103 98.8 — 101 99.8 — — Week 8 87 90.3 −8.2 (7.1) 87 94.2 −5.3 (7.3) 0.0037 Endpoint 100 90.6 −8.2 (7.1) 99 95.4 −4.5 (8.1) 0.0002 Evaluation of indapamide in elderly patients R Fiddes et al 242 Table 3 Summary of mean changes from baseline (week 0) to week 8 and endpoint in supine systolic blood pressure all-treated patients

Week Indapamide 1.25 mg Placebo P-value

n Mean value Mean change n Mean value Mean change (mm Hg) from baseline (mm Hg) from baseline (standard (standard deviation) deviation)

Week 0 10 159.3 — 101 160.3 — — 3 Week 8 87 147.9 −10.1 (13.8) 87 154.8 −4.2 (11.5) 0.0028 Endpoint 10 148.7 −10.5 (13.2) 99 157.2 −3.2 (13.2) 0.0002 0

Table 4 Summary of treatment responders (defined by decrease Table 6 Adverse events reported by at least 2% of patients in in supine diastolic blood pressure to р90 mm Hg or a reduction either treatment group during the double-blind treatment period of at least 10 mm Hg) all-treated patients Body system/Adverse event Indapamide Placebo Week Number and percentage (%) of patients P-value 1.25 mg (n = 101) (n = 103) Indapamide 1.25 mg Placebo Number and percentage (%) of Number Treatment Number Treatment patients of successes of successes Body as a whole patients patients Headache 5 (4.9) 7 (6.9) Asthenia 2 (1.9) 2 (2.0) Pain 2 (1.9) 2 (2.0) Week 8 87 48 (55%) 87 32 (37%) 0.020 Pain back 0 (0.0) 2 (2.0) Endpoint 100 55 (55%) 99 35 (35%) 0.005 Cardiovascular Abnormal electrocardiogram 0 (0.0) 2 (2.0) Palpitations 1 (1.0) 2 (2.0) Table 5 Summary of treatment responders (defined by decrease Atrial fibrillation 0 (0.0) 2 (2.0) in supine systolic blood pressure to р140 mm Hg) all-treated Infarction myocardial 0 (0.0) 3 (3.0) patients Central and peripheral nervous Cramps, leg 3 (2.9) 3 (3.0) Week Number and percentage (%) of patients P-value Dizziness 4 (3.9) 2 (2.0) Indapamide 1.25 mg Placebo Gastrointestinal Flatulence 1 (1.0) 2 (2.0) Number Treatment Number Treatment Tooth disease 0 (0.0) 2 (2.0) of successes of successes Metabolic and nutritional patients patients Oedema, peripheral 2 (1.9) 2 (2.0) Musculoskeletal Week 8 87 28 (32%) 87 12 (14%) 0.002 Arthritis 1 (1.0) 3 (3.0) Endpoint 100 30 (30%) 99 12 (12%) 0.001 Psychiatric Anxiety 1 (1.0) 3 (3.0) Respiratory Slightly larger mean changes in laboratory para- Pharyngitis 4 (3.9) 0 (0.0) meters from the double-blind baseline to endpoint Rhinitis 1 (1.0) 3 (3.0) (last on-study evaluation) were noted among Sinusitis 0 (0.0) 2 (2.0) patients in the indapamide 1.25 mg group compared Skin and appendages to patients in the placebo group, however none of Accidental Injury 0 (0.0) 2 (2.0) Rash 2 (1.9) 2 (2.0) the changes were considered to be clinically mean- ingful (Table 7). With respect to changes in laboratory parameters relative to the normal range, the majority of patients in both treatment groups had Ten of 100 (10%) patients with both baseline and laboratory values that were within the normal range endpoint potassium values in the indapamide at both the baseline and endpoint evaluations. Thir- 1.25 mg treatment group were hypokalaemic at least teen indapamide and five placebo patients had trig- once during the double-blind treatment period. Of lyceride values that shifted from within the normal these 10 patients, two were hypokalaemic and eight range at baseline to above the normal range at were normokalaemic at endpoint. Neither of the two endpoint, and 13 indapamide and two placebo patients who were hypokalaemic at endpoint patients had uric acid values that shifted from received potassium supplementation. of the eight within the normal range at baseline to above the nor- patients who were normokalaemic at endpoint, mal range at endpoint. In addition, seven indapam- three required potassium supplementation and five ide and one placebo patient had HDL cholesterol returned to normal levels without potassium sup- values that shifted from within the normal range at plements. The effect of potassium supplementation baseline to below the normal range at endpoint. on mean potassium values was minimal. None of the Evaluation of indapamide in elderly patients R Fiddes et al 243 Table 7 Mean changes (± standard deviation) in selected labora- changes in serum lipids and effects on serum potass- tory parameters from baseline (week 0) to the last on-study obser- ium and insulin resistance. Few data indicate that vation (endpoint)a these changes are of clinical significance or that they 7 Parameter Indapamide Placebo adversely affect cardiovascular risk. 1.25 mg (n = 93–100)b In several recently conducted long-term clinical (n = 95–103)b trials with high-dose diuretics,8 two of which were conducted in elderly hypertensive patients, no more Glucose (mg/dL) than a minimal increase in cholesterol levels was Baseline mean 106.4 104.1 + − reported; most trials reported a decrease. Further- Mean change 3.5 (27.2) 1.0 (20.7) more, the incidence of diabetes increased by only Potassium (mEq/L) approximately 1% in these patients. Therefore, the Baseline mean 4.3 4.3 Mean change −0.3 (0.4) +0.1 (0.3) use of these drugs in patients with hyperlipidaemia or diabetes should not be of much concern. In Triglycerides (mg/dL) Baseline mean 182.0 153.1 addition, hypokalaemia, another of the side effects Mean change +14.7 (69.1) −5.5 (59.9) associated with thiazide diuretics, is clearly dose- related and not a major problem with dosages Total Cholesterol (mg/dL) 8 Baseline mean 216.9 209.3 presently suggested. Although metabolic alterations Mean change +6.1 (21.7) +3.4 (19.7) can be minimized by lowering the dose, thiazide HDL Cholesterol (mg/dL) diuretics should be used cautiously in patients with Baseline mean 45.0 44.7 gout (due to the potential for hyperuricaemia) and Mean change −0.8 (6.2) +1.0 (6.7) in patients with diabetes mellitus (due to the poten- LDL Cholesterol (mg/dL) tial for worsened glucose tolerance). Baseline mean 143.5 140.1 The Joint National Committee has recommended + + Mean change 4.3 (22.0) 3.5 (18.2) that antihypertensive therapy be initiated at lower Uric acid (mg/dL) drug doses.9 In a recently published study, indapam- Baseline mean 5.7 5.8 ide 1.25 mg was shown to be safe and effective in Mean change +0.7 (0.9) −0.0 (0.8) treating hypertension in a slightly elderly patient Calcium (mg/dL) population (50 years of age or older).10 As compared Baseline mean 8.9 9.0 Mean change +0.0 (0.4) +0.1 (0.5) to placebo, indapamide 1.25 mg daily produced significant reductions in diastolic and SBPs, while sig- Sodium (mEq/L) Baseline mean 140.0 140.4 nificant differences in metabolic parameters were Mean change −0.3 (2.4) −0.2 (2.5) not observed. Hyperuricaemia and hypokalaemia Creatinine (mg/dL) occurred in less than 2% of indapamide-treated Baseline mean 1.2 1.2 patients. Mean change +0.0 (0.1) 0.0 (0.1) The present study was designed to evaluate the BUN (mg/dL) efficacy and safety of low-dose indapamide Baseline mean 15.4 15.9 (1.25 mg) in a true elderly patient population (65 Mean change +1.5 (4.2) −0.2 (3.8) years of age or older) with mild to moderate hyper- Magnesium (mEq/L) tension. Results showed indapamide 1.25 mg to be Baseline mean 1.7 1.7 − − superior to placebo in reducing supine diastolic and Mean change 0.0 (0.2) 0.0 (0.1) SBPs as well as standing diastolic and SBPs after 8 weeks of treatment. aAll analyses were summarized using descriptive statistics; no formal test of hypothesis was carried out. In terms of incidence of hypokalaemia (defined as bPatients with data at baseline or endpoint are included in the a potassium level р3.4 mEq/L), none of the placebo- analysis by timepoint. Patients with data at baseline and endpoint treated patients and 10% of the indapamide-treated are included in the analysis of mean changes from baseline. patients were hypokalaemic at least once during the double-blind treatment period. Only two of the 98 placebo-treated patients were hypokalaemic dur- patients with hypokalaemic occurrences were still ing the double-blind treatment interval. hypokalemic at endpoint; neither patient received potassium supplementation. Discussion For patients with arrhythmias or underlying ischaemic disease, a reduction in potassium loss Treatment of hypertension in the elderly patient could be potentially of clinical importance.11,12 population presents challenges, in part due to the However, studies that used 24- and 48-hour ECG presence of concomitant diseases, including heart monitoring do not support this assumption, as there failure, chronic obstructive pulmonary disease, dia- has been no consistent evidence of increased betes, gout, peripheral vascular disease, and renal arrhythmia associated with thiazide-induced hypo- failure. Several treatment options are available for kalaemia.13,14 Hypokalaemia has been associated the elderly hypertensive patient population, includ- with the occurrence of ventricular arrhythmias during treatment with diuretic-based regimens. The ing myocardial infarction.15,16 Other patient popu- efficacy of thiazide diuretics in reducing BP in the lations that could benefit from therapy that has a elderly patient population has been supported by minimal effect on electrolytes are those with recent the results of several clinical trials.1–5 Although thia- myocardial infarctions or patients on digitalis ther- zides are generally well-tolerated in the elderly,1,2,6 apy.17,18 In addition, the 1.25 mg dose may be parti- there is evidence that they produce short-term cularly suited to elderly patients who are generally Evaluation of indapamide in elderly patients R Fiddes et al 244 sensitive to volume depletion and who frequently European Working Party on High Blood Pressure in the have impaired cardiovascular reflexes that render Elderly. J Hypertens 1991; 9: 225–230. these patients more susceptible to hypotension. 7 Moser M. Current hypertension management: separat- Because of the presence of concomitant disease ing fact from fiction. Cleveland Clinic J Med 1993; 60 processes, this study was defined by the importance (1): 27–37. 8 Moser M. Hypertension. In: R Rakel (ed). Conn Current of administering low-dose indapamide to elderly Therapy. WB Saunders Company: Philadelphia PA, patients with mild to moderate hypertension. Low- 1995, pp 263–280. dose indapamide, 1.25 mg, given once-daily for 8 9 The Joint National Committee. The Fifth Report of the weeks was superior to placebo with respect to BP Joint National Committee on Detection, Evaluation and reduction as well as being safe and generally well- Treatment of High Blood Pressure. (JNC V) Arch Intern tolerated in this patient population. Med 1993; 153: 154–183. 10 Weidler D et al. Efficacious response with lower dose indapamide therapy in the treatment of elderly Acknowledgements patients with mild to moderate hypertension. J Clin This study was supported by research funding of Pharm 1995; 35: 45–51. 11 Hollifield JW. Potassium and magnesium abnormali- Rhoˆ ne-Poulenc Rorer Pharmaceuticals, Inc. ties: diuretics and arrhythmias in hypertension. Am J Med 1984; 77 (Suppl 5A): 28–32. References 12 Steward DE et al. Arrhythmogenic potential of diuretic-induced hypokalemia in patients with mild 1 SHEP Cooperative Research Group. Prevention of hypertension and ischemic heart disease. Acta Med stroke by antihypertensive drug treatment in older per- Scand 1988; 320: 1150–1177. sons with isolated systolic hypertension: final results 13 Papademetriou V et al. Thiazide therapy is not the of the Systolic Hypertension in the Elderly Program cause of arrhythmia in patients with systemic hyper- (SHEP). JAMA 1991; 265 (24): 3255–3264. tension. Arch Intern Med 1988; 148: 1272–1276. 2 MRC Working Party. Medical Research Council trial of 14 Madias JE, Madias NE, Gavras HP. Nonarrhythmogen- treatment of hypertension in older adults: principal icity of diuretic-induced hypokalemia. Arch Intern results. Br Med J 1992; 304: 405–412. Med 1984; 144: 2171–2176. 3 Amery A et al. Mortality and morbidity results from 15 Nordrehaug JE, Johannessen K-A, von der Lippe G. the European Working Party on High Blood Pressure Serum potassium concentration as a risk factor of ven- in the Elderly Trial. Lancet 1985; 1: 1349–1354. tricular arrhythmias early in acute myocardial infarc- 4 Management Committee. Treatment of mild hyperten- tion. Circulation 1985; 71 (4): 645–649. sion in the elderly. A study initiated and administered 16 Nordrehaug JE. Hypokalemia, arrhythmias and early by the National Heart Foundation of Australia. Med J prognosis in acute myocardial infarction. Acta Med Aust 1981; 2: 398–402. Scan 1985; 217: 299–306. 5 Materson BJ et al. Treatment of hypertension in the 17 Schulman M, Narins RG. Hypokalemia and cardio- elderly: I. Blood pressure and clinical changes. Results vascular disease. Am J Cardiol 1990; 65: 4E-9E. of a Department of Veterans Affairs Cooperative Study. 18 Clausen TG, Brocks K, Ibsen H. Hypokalemia and ven- Hypertension 1990; 15: 348–360. tricular arrhythmias in acute myocardial infarction. 6 Fletcher A et al. Risks and benefits in the trial of the Acta Med Scand 1988; 224: 531–537.