Nebivolol Vs Enalapril in the Treatment of Essential Hypertension: a Double-Blind Randomised Trial

ORIGINAL ARTICLE Nebivolol vs enalapril in the treatment of essential hypertension: a double-blind randomised trial

L Van Nueten1, A Schelling2, C Vertommen1, AG Dupont1 and JIS Robertson3 1Janssen Research Foundation, Beerse, Belgium; 2St Fransiscus Gasthuis, Rotterdam, The Netherlands; 3Glasgow, UK

-The trough-to .(0.002 ؍ The efficacy and acceptability of nebivolol 5 mg and response rate (70% vs 55%; P enalapril 10 mg, each given once daily, were compared peak sitting diastolic ratios also favoured nebivolol ,Nebivolol, but not enalapril .(0.002 ؍ in essential hypertension in a multicentre, randomised, (84% vs 60%, P double-blind trial over 3 months. For the index pre- slightly but significantly lowered heart rate. Both drugs declared variable, sitting diastolic pressure at trough were well-tolerated, although enalapril was drug level, nebivolol achieved greater falls in pressure accompanied by a significantly higher incidence of .and a higher coughing (0.009 ؍ ؊12.3 vs ؊9.9 mm Hg; P)

Keywords: vasodilating ␤-blockers; therapeutic trials; ACE inhibition; nitric oxide

Introduction Patients eligible at screening were informed of the trial procedures, and those entering were required ␤ Nebivolol is a highly cardioselective -blocker with to give informed consent. These subjects were then vasodilating properties; it acts in part via the L- given single-blind placebo for 1 month. Any pre- arginine/nitric oxide pathway; it is devoid of intrin- 1–9 sic sympathomimetic activity. Nebivolol has been Table 1 Exclusion criteria shown in human essential hypertension to control blood pressure (BP) effectively over 24 h in a dose of ¼ secondary hypertension; 5 mg once daily; no additional effect was seen with ¼ malignant hypertension (retinal haemorrhage, exudates or 10 mg.10–12 At similar antihypertensive doses to papillary oedema); ␤ ¼ asthma or chronic obstructive airway disease; established -blockers such as atenolol, nebivolol ¼ Ͻ ␤ bradycardia 60 beats/minute at rest; caused less -blockade as assessed by reduction of ¼ atrial fibrillation or tachyarrhythmia requiring exercise-induced tachycardia.9,13 Nebivolol antiarrhythmic therapy; improved left ventricular function when given to ¼ sick sinus syndrome or AV block greater than first degree; ¼ heart failure requiring treatment; patients with cardiac impairment and was well tol- ¼ 14–18 valvular disease of haemodynamic significance; erated in heart failure. The present study com- ¼ myocardial infarction or cerebrovascular accident within the prises a double-blind randomised parallel-group last 6 months; comparison of nebivolol with the angiotensin- ¼ insulin-dependent diabetes; converting enzyme (ACE) inhibitor enalapril in ¼ sensitivity or significant adverse reaction to beta-blockers or essential hypertension. ACE-inhibitors; ¼ significant renal (urine protein Ͼ trace, creatinine Ͼ2.2 mg/dl or Ͼ200 ␮mol/l) or hepatic (alamine-amino transferase and/or aspartate-amino transferase Ͼ2× the Patients and methods upper normal limit, total bilirubin Ͼ1.5× the upper normal The trial was conducted in 30 centres variously in limit) disease; ¼ renal artery disease; Argentina, Brazil, Denmark, France, Italy, Mexico, ¼ antecedents of auto-immune disease (leukopenia The Netherlands, South Africa, Sweden, and the Ͻ3500/mm3 and/or neutropenia Ͻ100 mm3); United Kingdom. Ambulatory patients, previously ¼ pregnancy, nursing or childbearing potential; treated or untreated, with a fifth phase diastolic ¼ any condition that could compromise the trial (alcohol or pressure over 94 mm Hg, and aged 18–74 years, were drug abuse, disabling illness, etc); ¼ concomitant medication that could affect blood pressure (eg, recruited. Exclusion criteria are listed in Table 1. tricyclic antidepressants, monoamine oxidase inhibitors; corticosteroids; non-steroidal anti-inflammatory drugs); ¼ 50% over ideal weight, based on Body Mass Index on Correspondence: Dr L Van Nueten, Janssen Research Foundation, Metropolitan Life Insurance Company’s 1983 Height and International Clinical R&D, Internal Medicine, Turnhoutseweg Weight Table; ¼ 30, 2340 Beerse, Belgium investigational drug treatment within the past 30 days; ¼ Received 20 January 1997; revised 19 July 1997; accepted 5 predictable lack of co-operation. August 1997 Nebivolol vs enalapril in hypertension L Van Nueten et al 814 vious antihypertensive drugs were either stopped with antihypertensive drugs (eg, thirst, cold fingers immediately, or were gradually withdrawn over a and toes, flushing, coughing etc). maximum of 2 weeks. Patients attended the clinic At week 2 of the run-in period, and at the end of at 2 and 4 weeks of the placebo period. Those with the double-blind comparison, routine urine testing a diastolic BP (DBP) above 114 mm Hg at 2 weeks was done, and blood samples were taken for routine were allocated to active double-blind therapy forth- haematological and biochemical tests; whilst in a with. Untreated patients whose diastolic was below subset of centres, from a total of 154 patients, a 95 mm Hg at the 4th week of placebo did not pro- further 10 ml of blood was collected for more ceed further. Trial subjects were randomly assigned detailed analysis of lipids (total cholesterol, high- in blocks of 12 to receive orally, once-daily and density and low-density lipoproteins (HDL and double-blind, either nebivolol 5 mg or enalapril LDL) and triglycerides). 10 mg. Visits during this comparison period were Twelve-lead electrocardiography was performed scheduled for weeks 2, 4, 8 and 12. After 3 months at week 2 of the run-in period, at the end of the of double-blind therapy the patients in both treat- double-blind comparison, and at the end of the run- ment groups were reallocated randomly to a ‘run- out period. out’ period of 1 month, double-blind on either pla- Patients were free to discontinue the trial at any cebo or a continuation of their existing nebivolol or time. They could also be withdrawn for any reason enalapril therapy. Visits were scheduled at weeks 2 at the discretion of the investigator. They were to be and 4 of the run-out period. withdrawn if the sitting diastolic pressure exceeded Medication comprised tablets identical in size, 119 mm Hg after 2 weeks, or 114 mm Hg after 4 colour, and taste, containing either placebo, nebivo- weeks of double-blind treatment, or if the treatment lol 5 mg, or enalapril 10 mg. Throughout, one tablet code was broken for any reason. The trial was was to be taken once daily at breakfast, except when approved by the ethical supervisory committee of the subjects attended for BP assessment, when medi- each participating centre. cation was omitted until this had been completed. At the randomisation (‘baseline’) and at the com- Statistics pletion of the 3 months comparison period, BP was also measured 2–3 h after drug ingestion, ie, at times The pre-declared index variable to be assessed was likely to be close to peak plasma levels.19–22 Com- sitting diastolic pressure at trough drug level. The pliance was assessed by tablet count at the end of shift from baseline of sitting DBP at trough level was the comparison period. Systolic and diastolic BPs defined as the primary measurement. Assuming a (respectively taking Korotkoff phases 1 and 5) were difference of at least 4 mm Hg with a standard devi- measured at all scheduled visits employing a stan- ation of 12 mm Hg in this measurement between the dard sphygmomanometer. The cuff was required to two drug groups, with 80% power and 5% two- be at least 30 cm long and 13 cm wide; if the arm tailed significance, at least 142 patients were needed circumference exceeded 34 cm, an appropriate per treatment group. Therefore a minimum of 320 larger cuff was used. In each patient such measure- patients, 160 per group, were planned to be entered. ments were made by the same investigator (or mem- The statistical analysis was performed according ber of the investigator’s staff) in the same room at to the intent-to-treat principle, ie, on all patient’s the same time of day. Three consecutive sitting randomised, regardless of their compliance with the pressures were taken after the patient had rested for protocol. All statistical tests reported are two-tailed at least 5 min; the last value was used for decision- and a P-value р0.05 was considered significant, making and for statistical analysis, except for sus- unless specified otherwise. If for any reason a pected protocol deviations, where all readings were patient failed to complete the trial as planned, the considered. BP was further measured once after the last available data were taken as ‘end-point’. The patient had stood for 2 min. At the first visit BP was end-point evaluation was considered as the primary determined in both arms; if a difference of more than time point. 4 mm Hg was found, the arm with the higher press- Response to treatment was defined as a decrease ure was subsequently always used, otherwise the in sitting trough DBP to 90 mm Hg or below, or a right arm was employed throughout. decrease in diastolic pressure of at least 10 mm Hg Heart rate was recorded immediately after the vs baseline, if the treated diastolic values remained measurement of both sitting and standing BP on all above 90 mm Hg. Odds ratios were calculated to occasions. Patients were weighed, partially clothed, compare the response rates of the nebivolol group at each visit. with the enalapril group; additionally, ␹2 tests were Any adverse events were recorded either if men- performed on the number of responders. Trough-to- tioned spontaneously by the patient or in response peak ratio was calculated for the sitting DBP as fol- to the standard question, ‘Has treatment upset you lows:19 in any way?’ Positive events were those recorded in trough/peak = ⌬ DBP trough/⌬ DBP peak response to the further standard question, ‘Did you experience any beneficial effect from treatment?’ where ⌬ DBP = shift in DBP from baseline (ie, at Symptoms were additionally assessed at the end randomisation) to week 12 of the double-blind treat- of the run-in period, and at the end of the double- ment. blind period comparison period, using a standard The individual items of the symptom question- questionnaire10 which rated increasing severity of naire and the total symptom severity score, defined 35 symptoms commonly occurring during treatment as sum of all items, were presented descriptively. Nebivolol vs enalapril in hypertension L Van Nueten et al 815 Results vs the enalapril group. By the end of the comparison phase, mean values were 92.3 and 95.6 mm Hg on Patient characteristics nebivolol and enalapril respectively. Despite the Four hundred and nineteen patients entered the trial slightly lower initial value for the nebivolol group, proper, that is substantially more than the planned shifts from baseline to the end of the comparison minimum. Of these 208 were randomised to nebivo- period significantly favoured nebivolol (−12.3 lol and 211 to enalapril. Their characteristics, which mm Hg vs −9.9 mm Hg; P = 0.009) as being the more are detailed in Table 2, were broadly similar effective drug. The percentages of responders between the two treatment groups. (defined as in the Methods section) at week 12 of the comparison were 70% on nebivolol vs 55% on enalapril (P = 0.002). Patient drop-outs The trough-to-peak ratios for sitting diastolic In total, 65 patients did not complete the trial (29 pressure at 3 months of double-blind treatment were nebivolol and 36 enalapril; 54 in the comparison 83.9% for nebivolol and 60.1% for enalapril period and 11 in the run-out phase). During the com- (P = 0.02). Neither drug had an appreciable ortho- parison period the main reason for withdrawal was static hypotensive effect. inefficacy (six nebivolol and 17 enalapril) or adverse Thus with the present design, and with the doses events (nine nebivolol and 12 enalapril). During the employed, nebivolol performed rather better than run-out phase, the main reason for dropping out was enalapril in lowering BP. The superior trough-to- an intercurrent event. peak ratio indicated that once daily nebivolol gave more even control of BP over 24 h. With neither drug was the antihypertensive response influenced Compliance with therapy by age or gender. By tablet count, all except 11 patients took more than 80% of their medication during the run-in; during the comparison period, all except seven nebivo- Heart rate lol and three enalapril patients took more than 80%; and during the run-out 19 patients took less than Heart rates were similar in the two treatment groups 80% (six remaining on nebivolol, three randomised at baseline. Throughout the periods of active ther- from nebivolol to placebo, six remaining on enalap- apy, heart rate was significantly lower with nebivo- ril, and four randomised from enalapril to placebo). lol than enalapril, the respective mean ± s.e.m. values at the 12th week of comparison, sitting, being 68.7 ± 0.69 vs 74.5 ± 0.81; P Ͻ 0.001. Blood pressure Details of the trough values for sitting and standing systolic and diastolic BPs during the comparison Body weight and run-out phases are given in Figures 1 and 2. Both drugs substantially lowered arterial pressure. Neither during the run-in period nor during the The primary declared criterion of evaluation was comparison were any significant differences in body the sitting diastolic pressure at trough drug level. At weight seen between nebivolol and enalapril, the baseline, this value was slightly but significantly respective means remaining between 74.8 and lower (mean 0.9 mm Hg, P = 0.032) in the nebivolol 75.8 kg.

Table 2 Patient and disease characteristics

Nebivolol Enalapril (n = 208) (n = 211)

Sex: M/F, n 113/95 124/87 Age: median (min–max), yearsa 54 (19–74) 53 (27–71) Weight: median (min–max), kg 74.5 (41–109) 74 (41–126) Height: median (min–max), cm 168 (142–190) 170 (142–190) Body mass index, median (min–max), kg/m2 26.5 (18– 45) 26.0 (19–36) Race: black/Caucasian/Hispanic/mixed/oriental/missing, n 18/166/16/0/7/1 19/169/14/4/5/0 Low salt diet, n 67 75 Alcohol consumption, n 100 107 Smoking, n 53 47 Previous antihypertensive treatment, n 159 167 Severity of hypertension (sitting DBP, trough), n ¼ DBP Ͻ95 mm Hg 6 1 ¼ mild (95 р DBP Ͻ100 mm Hg) 46 41 ¼ moderate (100 р DBP Ͻ115 mm Hg) 121 129 ¼ severe (DBP у 115 mm Hg) 34 38 ¼ missingb 12

No appreciable differences in patient characteristics between the treatments at baseline. aData on age, weight, height and low salt diet not available in some patients (no more than six patients in either group). bThree patients only have data at the start of the trial, not at baseline. Nebivolol vs enalapril in hypertension L Van Nueten et al 816

Figure 1 Diastolic (DBP) and systolic (SBP) blood pressure, trough: comparison period (means ± s.e.m). W = week of double-blind comparison. Nebivolol (——), enalapril (----).

Symptom assessment nebivolol and seven enalapril). During the comparison period nine nebivolol and 12 enalapril patients From baseline to the end of the comparison period, stopped treatment permanently because of adverse the mean total symptom score did not deteriorate in events. either group, nor did the mean scores deteriorate for any of the individual symptoms. In most patients there were about as many improvements as deterio- Routine laboratory tests rations. Only for cough, and then only with enalapril, did deteriorations (n = 48) outnumber improve- Throughout, no consistent or relevant changes were ments (n = 18) during the comparison period in seen in any of the routine urinary, haematological, more than 10% of the patients per group. or biochemical tests.

Positive events Plasma lipids The incidence of positive experiences was similar As mentioned above, these data were obtained in a in both treatment groups, being reported by 53% of subset of patients (80 nebivolol and 74 enalapril). the nebivolol and 55% of the enalapril subjects dur- The plasma levels of total cholesterol, high and low ing the comparison period. For both groups, the density lipoproteins (HDL/LDL) remained constant most frequently mentioned positive experience was throughout. For plasma triglycerides the nebivolol enhanced general well-being. group showed a rise in the mean concentration from 164.1 mg/dl at the end of the run-in to 194.3 mg/dl at the end of the 12 weeks comparison period, vs a Adverse events fall from 162.3 to 149.1 mg/dl in the enalapril group. The total of reported adverse experiences in at least These differences were due to marked increases ﬁve patients throughout the double-blind period of occurring in just two patients in the nebivolol group, the trial is shown in Table 3. respectively from 370 to 1067 and from 170 to 695 The adverse events most often reported during the mg/dl. When these outliers are omitted, the trigly- double-blind comparison period were headache (24 cerides levels in the nebivolol group rose from 161.4 nebivolol and 25 enalapril), cough (six nebivolol mg/dl at baseline to 175 mg/dl at the end of the com- and 21 enalapril; P = 0.0045), and paraesthesiae (13 parison period. Nebivolol vs enalapril in hypertension L Van Nueten et al 817

Figure 2 Diastolic (DBP) and systolic (SBP) blood pressure, trough: run-out period (mean ± s.e.m). W = week of comparison. Nebivolol/nebivolol (——), nebivolol/placebo (— - - —), enalapril/enalapril (----), enalapril/placebo (------).

Table 3 Adverse experiences (AE) reported in at least ﬁve Discussion patients

Nebivolol Enalapril The present trial has shown that nebivolol given (n = 208) (n = 211) once daily compares favourably with once-daily dosage of an established agent, the ACE inhibitor Headache 24 25 enalapril, in the treatment of essential hypertension. Paraesthesiae 13 7 The definitive dose of nebivolol (5 mg), was superior Nightmares 7 2 to 10 mg of enalapril in respect of the predeclared Palpitations 7 5 Cough 6 21 index measurement, the sitting diastolic pressure at Fatigue 6 4 trough drug level, both concerning the extent of BP Diarrhoea/nausea 6 6 reduction and the response rate. Chest pain 6 5 There was a considerably higher calculated Dizziness 5 7 trough-to-peak BP ratio with nebivolol than with Vertigo 5 5 Rash 2 6 enalapril. Peak plasma concentrations of nebivolol Impotence/libido decreased 2 7 occur on average around 1–1.5 hours after oral Percent of patients with AE 48.6% 55.0% ingestion in extensive metabolizers of the drug, who predominate in the population.20 Peak plasma levels of enalaprilat, the active metabolite of enalapril, are found around 4 h after ingestion.21,22 It is thus poss- ible that with the assessment time chosen after drug ingestion, the peak plasma levels had slightly Electrocardiogram passed with nebivolol, and, by contrast had not been No unexpected systematic changes were observed in achieved with enalaprilat. However, as mentioned, the electrocardiogram. At the end of the comparison nebivolol showed the better antihypertensive effect period, in comparison with baseline, the nebivolol plus more numerous responders at trough. Thus any group showed a decrease in heart rate, QRS, and tendency to underestimate the true peak effect of QTc and an increase in PQ and QT; while in the enalapril would have favoured a higher calculated enalapril group heart rate and QTc decreased. trough-to-peak ratio with that drug. Therefore the Nebivolol vs enalapril in hypertension L Van Nueten et al 818 superior trough-to-peak ratio of nebivolol appears to 7 Cockroft JR et al. Nebivolol vasodilates human forearm be a genuine finding. Lipicky23 has stated that 50– vasculature: Evidence for an L-arginine/NO-dependent 75% of the peak effect should be preserved at mechanism. J Pharmacol Exp Ther 1995; 274: 1067– trough; the present finding of 84% with nebivolol 1071. comfortably betters that criterion. While both drugs 8 Bowman AJ, Chen CPL-H, Ford GA. Nitric oxide mediated venodilator effects of nebivolol. Br J Clin were well tolerated, there was the expected excess 24 Pharmacol 1994; 38: 199–204. of coughing with enalapril. 9 Van Rooy P, Van Nueten L, De Creé J. Nebivolol blood Two individuals on nebivolol showed distinct pressure reduction in relation to beta-blockade in heal- increases in plasma triglycerides; otherwise no thy volunteers. Drug Invest 1991; 3 (Suppl 1): 174– appreciable changes in plasma lipids occurred. The 175. increases of triglycerides in these two individuals 10 Van Nueten L et al. A dose-response trial of nebivolol are surprising and unexplained. In studies reported in essential hypertension. J Hum Hypertens 1997; 11: previously11,25 no increases were found in plasma 139–144. total cholesterol, triglycerides, HDL, LDL, VLDL, or 11 Lacourcie`re Y et al. Treatment of ambulatory hyper- apolipoproteins A1 or B in patients treated for 3 tensives with nebivolol or hydrochorothiazide alone and in combination. A randomized, double-blind, pla- months with once-daily nebivolol 1 mg (n = 20), = = cebo-controlled, factorial-design trial. Am J Hypertens 5mg(n 46), or 10 mg (n 20). 1994; 7: 137–145. Beta-blocking drugs are now well-established in 12 Van Bortel LMAB et al. Nebivolol in hypertension: a the treatment of essential hypertension and have double-blind placebo-controlled multicentre study been employed in several prospective trials which assessing its antihypertensive efficacy and impact on have shown clear reductions in morbidity.26,27 Nebi- quality of life. J Cardiovasc Pharmacol 1993; 21: volol, a highly cardioselective ␤-blocker without 856–862. intrinsic sympathomimetic effect, and with vasodil- 13 Van Bortel LMAB et al. Pharmacological properties of ating properties, offers several attractions.1–5 Nebiv- nebivolol in man. Eur J Clinical Pharmacol 1997; 51: olol induces endothelium-dependent vascular relax- 379–384. 6–8 14 Stoleru I et al. Effects of d-nebivolol and l-nebivolol ation via the L-arginine/nitric oxide system. It on the left ventricular systolic and diastolic function: improves large arterial compliance in hyperten- comparison with d-l nebivolol and atenolol. J Cardiov- 28,29 sion. At similar antihypertensive doses, nebivo- asc Pharmacol 1993; 22: 183–190. lol causes less ␤-blockade than do classical ␤-block- 15 Goldstein M et al. Administration of nebivolol after ers, as assessed by the reduction of exercise-induced coronary bypass graft in patients with altered left ven- tachycardia.9 Nebivolol has been shown to be well- tricular function. J Cardiovasc Pharmacol 1993; 22: tolerated in cardiac failure, and to improve left ven- 253–258. tricular function in patients with cardiac impair- 16 Wisenbaugh T et al. Long-term (3 month) effects of a ment.14–18 new beta-blocker (nebivolol) on cardiac performance This study has amply confirmed the antihyperten- in dilated cardiomyopathy. J Am Coll Cardiol 1993; 21: sive efficacy, long duration of action, and excellent 1094–1100. acceptability of nebivolol. 17 Uhlir O et al. Nebivolol in the treatment of cardiac failure: a double-blind controlled clinical trial. In sub- mission to J Card Fail. Acknowledgements 18 Rousseau MF et al. Medium term effects of beta-blockade on left ventricular mechanics: a double-blind, pla- We are indebted to the trial monitors and coordi- cebo controlled comparison of nebivolol and atenolol nators in the different countries. We would also like in patients with ischemic left ventricular dysfunction. to thank Mrs Ilse Versmissen and Mrs Mies Sysmans J Card Fail 1996; 2: 15–23. for their secretarial assistance. 19 Rose M, McMahon F. Some problems with antihypertensive drug studies in the context of the new guide- lines. Am J Hypertens 1990; 3: 151–155. References 20 Van Peer A et al. Clinical pharmacokinetics of nebivo- 1 Fitzgerald JD (Ed). First International Nebivolol Inves- lol: a review. Drug Invest 1991; 3 (Suppl 1): 25–30. tigators’ Meeting. Drug Invest 1991; 3 (Suppl 1): 1–203. 21 Johnston CI. Angiotensin converting enzyme inhibi- 2 Van de Water A et al. Pharmacological and hemodyn- tors. In: Doyle AE (ed). Clinical Pharmacology of Anti- amic profile of nebivolol, a chemically novel, potent, hypertensive Drugs. Handbook of Hypertension, Vol. ␤ and selective 1-adrenergic antagonist. J Cardiovasc 11. Elsevier: Amsterdam, 1988, pp 301–326. Pharmacol 1988; 11: 552–653. 22 Hansson L et al. Angiotensin-converting enzyme 3 Pauwels PJ et al. The receptor binding profile of the inhibitors in the treatment of essential hypertension. new antihypertensive agent nebivolol and its stereo- In: Robertson JIS, Nicholls MG (eds). The Renin-Angi- ␤ isomers compared with various 1-adrenergic blockers. otensin System. Gower/Mosby: London/New York, Mol Pharmacol 1988; 34: 843–851. 1993, Vol. 2:91.1–91.24. 4 Janssens WJ et al. Nebivolol is devoid of intrinsic sym- 23 Lipicky RJ. Trough: peak ratio: the rationale behind pathomimetic activity. Eur J Pharmacol 1989; 159: the United States Food and Drug Administration rec- 89–95. ommendations. J Hypertens 1994; 12 (Suppl 8): 17–19. 5 Van de Water A, Xhonneux R, Reneman RS, Janssen 24 Fletcher AE, Dollery CT. Side effects associated with PAJ. Cardiovascular effects of dl-nebivolol and its inhibitors of angiotensin concerning enzyme. In: Rob- enantiomers—a comparison with those of atenolol. Eur ertson JIS, Nicholls MG (eds). The Renin-Angiotensin J Pharmacol 1988; 156: 95–103. System. Gower/Mosby: London/New York, 1993, 6 Gao Y et al. Nebivolol induces endothelium-depen- Vol. 2: 99.1–99.15. dent relaxations of canine coronary arteries. J Cardiov- 25 Lacourcie`re Y et al. Comparative effects of a new car- asc Pharmacol 1991; 17: 964–969. dioselective beta-blocker nebivolol and nifedipine sus- Nebivolol vs enalapril in hypertension L Van Nueten et al 819 tained-release on 24-hour ambulatory blood pressure Denmark and plasma lipoproteins. J Clin Pharm 1992; 32: Clemmensen P (Hilleroed), Gliese M (Glostrup), 660–666. Grande P (Hilleroed), Hansen EA (Haderslev), 26 Robertson JIS, Ball SG. Hypertension for the Clinician. Lassen E (Odder), Rishoj M (Roedovre), Wickers- WB Saunders: London/Philadelphia, 1994, pp 117– Nielsen N (Aalborg), Winkel OP (Slagelse). 118. 27 Robertson JIS. Clinical trials in hypertension. In: Pitt B, Julian D, Pocock S (eds). Clinical Trials in Cardi- France ology. WB Saunders: London/Philadelphia, 1997, pp Marcadet DM (Paris), Tcherdakoff Ph. (Boulogne). 115–160. 28 Van Bortel LMAB, Schiffers PMH, Mooij JMV. Invasive Italy haemodynamics of nebivolol in hypertensive patients. Celentano A, Galderisi M (Naples). Drug Invest 1991: 3 (Suppl 1): 60. 29 Van Merode T et al. Verapamil and nebivolol improve Mexico carotid artery distensibility in hypertensive patients. J Alcocer L, Bochicchio T (Mexico City). Hypertens 1989; 7 (Suppl 6): S262–S263. Netherlands Oldenbroek C (Hoorn), Schelling A, Schop C Appendix (Rotterdam). List of participating physicians South Africa Leary WP (Congella). Argentina Plotquin Y, Schenna RA, Senatore V (Buenos Aires). Sweden Jacobson B (Karlstad), Jaup B, Norrby A, Ulvenstam Brazil G (Goteborg), Ohlsson P (Motala). Oigman W (Rio de Janeiro), Ribeiro AB (Sao Paolo). United Kingdom Fairhurst GJ (St Helens), Silvert BD (Bolton).