Journal of Human Hypertension (1998) 12, 563–567  1998 Stockton Press. All rights reserved 0950-9240/98 $12.00 http://www.stockton-press.co.uk/jhh ORIGINAL ARTICLE Valsartan and atenolol in patients with severe essential hypertension

R Cifkova1, J Pelesˇka2, J Hradec3, H Rosolova´4, E Pinte´rova´5, K Zeman6, P Oddou-Stock7, J Thirlwell8 and F Botteri7 1Institute for Clinical and Experimental Medicine, Prague, Czech Republic; 2Medical Department II, General Teaching Hospital, Prague, Czech Republic; 3Medical Department III, General Teaching Hospital, Prague, Czech Republic; 4Medical Department II, Pilsen, Czech Republic; 5Medical Department II, Hradec Kra´ love´, Czech Republic; 6Medical Department I, St Anna’s Hospital, Brno, Czech Republic; 7Novartis Pharma AG, Basel, Switzerland; 8Novartis Pharmaceuticals, Horsham, UK

The aim of this study was to evaluate the efficacy and lol, ؊25.5 mm Hg. There was no statistically significant tolerability of valsartan, a new angiotensin II receptor difference between the treatment groups. Add-on hyd- antagonist, versus atenolol in the treatment of severe rochlorothiazide (HCTZ) 25 mg was required by 97.2% of primary hypertension. A total of 103 adult out-patients patients receiving atenolol and 83.6% of patients receiv- were randomised to receive either valsartan 160 mg or ing valsartan; additional verapamil SR 240 mg was also atenolol 100 mg once daily for 6 weeks. If necessary, required by 58.3% of patients receiving atenolol and additional blood pressure (BP) control could be pro- 64.2% receiving valsartan. Valsartan was well tolerated, vided as add-on therapy. Both valsartan and atenolol with a comparable incidence of treatment-related decreased mean sitting diastolic BP (DBP) and mean sit- adverse experiences in both groups. ting systolic BP (SBP): least squares mean change from In conclusion valsartan 160 mg is as well tolerated baseline in DBP; valsartan, ؊20.0 mm Hg; atenolol, and effective as atenolol 100 mg in lowering BP in .؊20.4 mm Hg: in SBP; valsartan, ؊30.0 mm Hg; ateno- severely hypertensive patients

Keywords: valsartan; angiotensin II receptor blocker; angiotensin II receptor antagonist; severe essential hypertension; atenolol

Introduction hypertension (diastolic BP [DBP] р115 mm Hg).8–11 The purpose of the current study was to evaluate the Primary hypertension can be controlled by modul- efficacy and tolerability of valsartan in the treatment ation of the renin-angiotensin system, which plays of patients with severe hypertension, and to com- a fundamental role in the regulation of the extra- 1 pare 160 mg valsartan with a reference anti-hyper- cellular fluid volume and blood pressure (BP). tensive therapy, 100 mg atenolol, both as initial Pharmacologically, this has been achieved by monotherapy and in combination with other anti- inhibiting the angiotensin-converting enzyme hypertensive agents. (ACE), but this treatment produces characteristic side effects related to inhibition of other ACE sub- strates.2–5 More recently, a novel class of anti-hyper- Patients and methods tensives has been developed, the angiotensin II Patients receptor antagonists, which specifically and com- 6 Men and women aged 25–68 with uncomplicated petitively block the AT1 receptor. Valsartan is a new, orally active, potent and highly selective severe primary hypertension (defined as a mean sit- angiotensin II receptor antagonist7 which at doses of ting DBP у110 and Ͻ120 mm Hg on randomisation 80 and 160 mg once daily has been shown to be day) were eligible for inclusion in the trial. Patients effective in the treatment of mild to moderate hyper- were either newly diagnosed or had their BP inad- tension compared with placebo8–10 and to be safe equately controlled by their previous treatment. The and well tolerated, lacking the side effects character- major exclusion criteria were: secondary hyperten- istic of the ACE inhibitors.11 sion, insulin-dependent diabetes mellitus, serious To date all published studies have evaluated the heart, liver or renal disease, history of cerebrovascu- use of valsartan in patients with mild to moderate lar accident, intracranial haemorrhage, hypertensive encephalopathy, serum potassium level Ͻ3.5 mmol/L, urinary tract infection, or clinically sig- nificant allergies. Fertile women were required to Correspondence: Dr Renata Cifkova, Institute for Clinical and Experimental Medicine, Prague, Czech Republic use effective contraception throughout the trial. Received 31 December 1997; revised 10 February 1998; accepted All patients gave written consent to participation 6 March 1998 in the trial, which was approved by a local Insti- Valsartan and atenolol in severe hypertension R Cifkova et al 564 tutional Review Board. The study was performed of patients whose BP was controlled (DBP Ͻ95 according to the revised Declaration of Helsinki and mm Hg), and the change from baseline in standing Good Clinical Practice requirements. Patients could systolic and diastolic BP at end-point. be withdrawn from the trial if they had intolerable The primary criterion for safety and tolerability adverse experiences or inadequate therapeutic was the reporting of adverse experiences (AEs). At response resulting in intolerable symptoms and/or each visit all AEs which had occurred since the pre- DBP Ͻ60 mm Hg or у120 mm Hg, for major vious visit were recorded, with particular attention violations of the protocol, significant non-com- paid to reports of orthostatic hypotension, as pre- pliance, hypokalaemia Ͻ3 mmol/L, or if it was felt viously defined,13 or dry cough.14 The secondary cri- to be in the patient’s best interest. teria for safety and tolerability were changes in body weight or heart rate, and any clinically relevant changes in routine haematological and biochemical Study design laboratory findings or 12-lead ECG. This was a multicentre, randomised, double-blind, active controlled, parallel trial performed in the Czech Republic. After an initial washout period of Statistical methodology 3 days, patients were given single-blind placebo for As the primary aim of the study was to compare the 5 days. This relatively short run-in period (8 days tolerability and safety profiles of valsartan and aten- in total) was due to the severity of the hypertensive olol, the intended sample size of 102 patients was disease in patients participating in the study. Eli- determined pragmatically to provide 60 patients gible patients were randomised into two groups to exposed to valsartan for at least 6 weeks and give a 2:1 ratio of valsartan to atenolol treatment, in allowing for a drop-out rate of 10%. It was calcu- order to maximise the data obtained from the valsar- lated that if 50% of atenolol-treated patients tan treatment group. Patients received either valsar- reported an AE, this sample size could detect an tan 160 mg (a dose which has been shown to be absolute difference of 30% in comparison with val- effective and safe in patients with mild to moderate sartan with 80% power at the 5% significance level. 11 hypertension) or atenolol 100 mg (a dose con- The intent-to-treat population, ie, all randomised sidered comparable to valsartan 160 mg), as two cap- patients with at least one post-baseline examination, sules (2 × valsartan 80 mg or 2 × atenolol 50 mg) to was used for the statistical analysis. be taken once daily at 8 am. Study drugs were sup- The changes in DBP and SBP, and the change from plied in identical capsules to maintain blinding. baseline in sitting heart rate and body weight, were Patients were assessed twice during the washout analysed using a two-way covariance model and single-blind placebo treatment periods, once at (ANCOVA)15 fitting baseline centre and treatment. randomisation (baseline measurement), and again Treatment-by-baseline and centre-by-treatment after 1, 2, 3, 4, 5 and 6 weeks of active treatment. interactions were tested for statistical significance. Scheduled assessment visits were made in the Response rates were analysed using a chi-square morning before the study medication for that day test.16 Analyses were performed at the two-sided 5% was taken, to provide trough BP measurements. significance level. The potential risk of leaving severely hypertensive The incidence of adverse experiences in each patients without adequate BP control justified the treatment group was compared using a chi-square use of add-on therapies. If necessary, hydrochloro- test. thiazide (HCTZ) 25 mg could be started 2 weeks after randomisation. If BP was still not controlled (a patient was considered to be controlled if the DBP Results Ͻ was 95 mm Hg) at 4 weeks, verapamil SR 240 mg Patients could be given in addition to HCTZ. Thus the data compare valsartan vs atenolol, in both cases with the A total of 103 hypertensive patients were random- optional addition of HCTZ and verapamil. The use ised into the two treatment groups: valsartan 160 mg of other medication which could have interfered (n = 67) and atenolol 100 mg (n = 36). There were no with the assessment of end-points was prohibited major differences between the treatment groups in for the duration of the trial. demographics or baseline characteristics (Table 1). The main efficacy variables were the change from All patients were Caucasian and the majority were baseline in trough DBP and mean sitting systolic BP male (70.9%), with proportionately slightly more (SBP) after 6 weeks of treatment or at the time of men in the atenolol (77.8%) than in the valsartan premature discontinuation. Systolic and diastolic group (67.2%). The median duration of exposure to BP was measured at each visit by the same clinician trial medication for both groups was 42 days. Add- in the same arm using the same sphygmomano- on HCTZ was required at some point by a higher meter, and all BP measurements were made using percentage of patients receiving atenolol (97.2%) WHO criteria12 to the nearest 2 mm Hg. For diastolic than valsartan (83.6%). However, the further pressure, phase V (disappearance of Korotkoff addition of verapamil was needed by a higher per- sounds) was used. centage of the valsartan group (64.2%, or 76.8% of Other efficacy variables included responder rates those already on HCTZ) than the atenolol group (ie, the number of patients who normalised their (58.3%, or 60% of those already on HCTZ). There DBP to Ͻ90 mm Hg or achieved a decrease in DBP was a total of seven premature discontinuations dur- of у10 mm Hg compared to baseline), the number ing the trial, with 96 (93%) patients completing the Valsartan and atenolol in severe hypertension R Cifkova et al 565 Table 1 Patient demographics and baseline data

Randomised patients Valsartan Atenolol 160 mg 100 mg

Number of randomised patients 67 36 Sex (%) Male 45 (67.2) 28 (77.8) Female 22 (32.8) 8 (22.2) Race (%) Caucasian 67 (100) 36 (100) Other 0 (0) 0 (0) Age (mean ± s.d.) 53 ± 8.2 52.6 ± 8.8 Duration of hypertension in years 9.6 9.8 (median) Body weight in kg (mean ± s.d.) 88.1 ± 11.3 89.1 ± 13.6 Baseline sitting DBP in mm Hg 113.5 ± 2.8 113.6 ± 3.0 (mean ± s.d.) Baseline sitting SBP in mm Hg 173.9 ± 17.2 174.3 ± 16.8 (mean ± s.d.) Sitting heart rate in beats/min 76.3 ± 8.6 77.2 ± 11.2 (mean ± s.d.) Significant medical 48 (71.6) 25 (69.4) history/concomitant diseases (%)

full 6 weeks of active double-blind treatment. Four patients (6%) discontinued from the valsartan group, three (8.3%) from the atenolol group. Most discontinuations were due to adverse experiences Figure 1 Mean sitting diastolic blood pressure (MSDBP, top) and mean sitting systolic blood pressure (MSSBP, bottom) at −8, 0, 7, (two of four in the valsartan group, all three in the 14, 21, 28, 35 and 42 days of treatment with valsartan 160 mg atenolol group). (n = 67) or atenolol 100 mg (n = 36) from the dataset of random- ised patients. Day 0 is day of randomisation. Efficacy The results shown are for the intent-to-treat dataset. similar (Figure 1) and indicated that DBP and SBP Both valsartan and atenolol markedly reduced mean were still declining at the end of the 6-week study sitting diastolic and systolic BPs at the 6-week end- period. No significant treatment-by-baseline or treat- point (least squares mean change from baseline in ment-by-centre interactions were seen. − − The response rate in the valsartan group (85.1%) DBP: valsartan, 20.0 mm Hg; atenolol, 20.4 = mm Hg; least squares mean change from baseline in was comparable (P 0.887) to that in the atenolol SBP: valsartan, −30.0 mm Hg; atenolol, −25.5 group (86.1%) (Table 3). Both groups were also com- parable in the number of controlled patients, mm Hg) (Table 2). There was no statistically signifi- р cant difference between the effect of valsartan defined as having a DBP 95 mm Hg at end-point 160 mg or atenolol 100 mg on either variable. The (valsartan 59.7%, atenolol 66.7%). The number who р profiles of action of both drugs in lowering BP were had mean DBP 90 mm Hg at end-point was 29 (43.3%) in the valsartan group and 20 (55.6%) in the atenolol group. The mean change from baseline at end-point in standing diastolic and systolic BP was Table 2 Comparison of valsartan and atenolol treatment groups comparable in both treatment groups (valsartan −17.4 and −27.2 mm Hg, atenolol −19.0 and −28.2 Valsartan Atenolol mm Hg, for diastolic and systolic BP respectively). (n = 67) (n = 36) Table 3 Rate of BP response and control at end-point DBP ± − ± − ± Mean change (mm Hg s.d.) 19.9 9.70 20.7 9.9 = = Statistical significance (P) 0.819 Valsartan (n 67) Atenolol (n 36)

SBP Responders 57 (85.1%) 31 (86.1%) Mean change (mm Hg ± s.d.) −30.7 ± 16.8 −26.5 ± 17.4 Non-responders 10 (14.9%) 5 (13.9%) Statistical significance (P) 0.161 P 0.887

No. (%) patients requiring additional 56 (83.6%) 35 (97.2%) Controlled 40 (59.7%) 24 (66.7%) HCTZ Not controlled 27 (40.3%) 12 (33.3%) No. (%) patients requiring additional 43 (64.2%) 21 (58.3%) + HCTZ verapamil Responders: patients achieving DBP Ͻ90 mm Hg or a decrease у10 mm Hg compared to baseline. DBP, mean sitting diastolic blood pressure. Controlled: patients achieving DBP Ͻ95 mm Hg. No formal stat- SBP, mean sitting systolic blood pressure. istical analysis was performed. Valsartan and atenolol in severe hypertension R Cifkova et al 566 Safety and tolerability orated by the similar responder and controlled BP rates in both groups. Both study medications were generally well toler- One of the main aims of the study was to acquire ated. A total of 44 patients (42.7%) reported one or more data on the systemic safety and tolerability of more AE, whether or not it was related to trial drug valsartan 160 mg. The results show that valsartan medication. Of these, 26 (38.8%) were in the valsar- and atenolol were both well tolerated. A total of 44 tan group and 18 (50%) in the atenolol group. Most of the 103 randomised patients reported adverse reported AEs were mild to moderate in severity. experiences, with more patients in the atenolol treat- Seventeen (25.4%) patients in the valsartan group ment group reporting AEs (50% vs 38.8%) although and nine (25.0%) patients in the atenolol group there was no significant difference between the reported AEs that the investigator considered to be groups in the incidence of AEs regardless of trial at least possibly related to trial medication. The drug relationship. The commonest AE in valsartan- most common drug-related AEs in the valsartan treated patients was fatigue, and in atenolol-treated group were fatigue (11.9%), headache (6.0%) and patients it was headache. Most AEs were considered dizziness (6.0%) (Table 4). The commonest drug- to be mild or moderate. No orthostatic hypotension related AE in patients receiving atenolol was head- was reported. One patient in each treatment group ache (11.1%), while fatigue, dyspnea and muscle had a serious AE leading to premature withdrawal weakness were each reported by two patients (5.6%) from the trial. in the atenolol group. Dry cough of mild severity Previous data showed that valsartan is effective and considered to be treatment-related was reported and safe in the treatment of mild to moderate hyper- by one patient (1.5%) in the valsartan group and by tension, but these studies had not been extended to none in the atenolol group. No patients reported severely hypertensive patients. The treatment of sev- orthostatic hypotension. ere hypertension often requires two or more differ- Only two patients had AEs considered to be seri- ent drugs, although monotherapy would ideally be ous and at least possibly drug-related. Two further preferable in terms of convenience and compliance. patients prematurely discontinued for reasons con- As expected, it was found that many patients with sidered probably related to trial treatment: one in severe primary hypertension in this study required the valsartan group for hypokalaemia after having anti-hypertensive therapy in addition to valsartan or received HCTZ as add-on therapy for 2 weeks, and atenolol. Thus the results compare valsartan and one in the atenolol group for mild dyspnea. atenolol against a background of HCTZ and verapa- The change in sitting heart rate showed a statisti- mil therapy. However, by the end of the trial eight cally significant difference between the valsartan (11.9%) patients were still on valsartan monother- and atenolol groups (least squares mean change in apy, compared to only one (2.8%) patient whose BP sitting heart rate from baseline on valsartan was −1.8 could be adequately controlled on atenolol alone, beats/min, on atenolol −7.3, P Ͻ 0.001). The results suggesting that valsartan may offer some advantage were similar for standing heart rate. There was no over atenolol in BP control. This might be expected evidence of a statistically significant effect on body in severe hypertension, where the renin-angiotensin weight or of any clinically relevant ECG or labora- system could play a major role. Direct comparison tory finding in either treatment group. Creatinine of these data with another angiotensin II receptor excretion rates in both groups were comparable antagonist, losartan, is difficult because the pub- (valsartan 160 mg: mean change from baseline lished treatment regimens differ, but indicates that 0.042 g/8h; atenolol 100 mg: −0.045 g/8h). the proportion of patients whose severe hyperten- sion was adequately controlled by valsartan mono- Discussion therapy is comparable to that reported for at least one trial using losartan.17 This is the first study evaluating the safety and effi- In summary, this study shows that valsartan, a cacy of valsartan 160 mg in patients with severe pri- new angiotensin II antagonist, is as effective and mary hypertension. The data suggest that valsartan well tolerated as atenolol, a first-line treatment in is as effective as the comparator, atenolol. Both treat- essential hypertension, in reducing BP in severely ments produced a clinically significant decrease hypertensive patients. The study suggests that the from baseline in DBP and SBP, with no significant angiotensin II receptor antagonist valsartan offers a difference observed between the two groups. The promising alternative to atenolol in the treatment of comparability of valsartan and atenolol was corrob- uncomplicated severe primary hypertension.

Acknowledgements Table 4 Incidence of most frequent (Ͼ3%) adverse experiences (AEs) considered drug related We would like to acknowledge the collaboration and commitment of all the local investigators and their = = Valsartan (n 67) Atenolol (n 36) staff without whom the present study would not have been possible. Fatigue 8 (11.9) 2 (5.6) Headache 4 (6.0) 4 (11.1) Dizziness 4 (6.0) 0 (0.0) References Dyspnea 0 (0.0) 2 (5.6) Muscle weakness 0 (0.0) 2 (5.6) 1 Sealey JE, Laragh JH. The renin-angiotensin-aldos- terone system for normal regulation of blood pressure Valsartan and atenolol in severe hypertension R Cifkova et al 567 and sodium and potassium homeostasis. In: Brenner antagonist for the treatment of essential hypertension: BM, Laragh JH (eds). Hypertension. Raven Press: New efficacy and safety compared with placebo and enalap- York, 1990, pp 1287–1319. ril. J Hypert 1996; 14: 1147–1151. 2 Sassano P et al. Antihypertensive effect of enalapril as 10 Hegner G et al. Valsartan, a new angiotensin II antag- first-step treatment of mild and moderate uncompli- onist for the treatment of essential hypertension: effi- cated essential hypertension: Evaluation by two cacy and safety compared with hydrochlorothiazide. methods of blood pressure measurement. Am J Med Eur J Clin Pharmacol 1997; 52: 173–177. 1984; 20: 18–22. 11 Oparil S et al. The efficacy and safety of valsartan com- 3 Frishman WH, Goldberger J, Sherman D. Enalapril, pared with placebo in the treatment of patients with hydrochlorothiazide, and combination therapy in essential hypertension. Clin Ther 1996; 18: 797–810. patients with moderate hypertension. J Clin Hypertens 12 Subcommittee of WHO/ISH Mild Hypertension Liai- 1987; 3: 520–527. son Committee. Summary of 1993 World Health 4 Kaufman MP, Coleridge HM, Coleridge JCG, Baker DG. Organisation-International Society of Hypertension Bradykinin stimulates afferent vagal C-fibers in intra- guidelines for the management of mild hypertension. pulmonary airways of dogs. Am Physiol Soc 1980; 48: Br Med J 1993; 307: 1541–1546. 511–517. 13 Lipsitz LA. Orthostatic hypotension in the elderly. 5 Israili ZH, Hall WD. Cough and angioneurotic edema N Engl J Med 1989; 321: 952–957. associated with angiotensin-converting enzyme inhibi- 14 Morice AH, Brown MJ, Higenbottam T. Cough associa- tor therapy. Ann Int Med 1992; 117: 234–242. ted with angiotensin converting enzyme inhibition. 6 Kang PM, Landau AJ, Eberhardt RT, Frishman WH. J Cardiovasc Pharmacol 1989; 13 (Suppl 3): S59-S62. Angiotensin II receptor antagonists: a new approach to 15 Armitage P, Berry G. The analysis of covariance. In: blockade of the renin-angiotensin system. Am Heart J Statistical Methods in Medical Research, Blackwell 1994; 127: 1388–1401. Scientific Publications, 1987, pp 282–295. 7Mu¨ ller P et al. Angiotensin II receptor blockade with 16 Armitage P, Berry G. 2 × 2 tables and chi-square tests. single doses of valsartan in healthy, normotensive sub- In: Statistical Methods in Medical Research. Blackwell jects. Eur J Clin Pharmacol 1994; 47: 231–245. Scientific Publications, 1987, pp 125–129. 8 Corea L et al. Valsartan, a new angiotensin II antagon- 17 Ruff D et al. Comparative effects of combination drug ist for the treatment of essential hypertension: A com- therapy regimens commencing with either losartan parative study of the efficacy and safety against amlo- potassium, an angiotensin II receptor antagonist, or pidine. Clin Pharm Ther 1996; 60: 341–346. enalapril maleate for the treatment of severe hyperten- 9 Holwerda NJ et al. Valsartan, a new angiotensin II sion. J Hypertens 1996; 14: 263–270.