- Email: [email protected]
A comparison of candesartan, felodipine, and their combination in the treatment of elderly patients with systolic hypertension
AJH
2002; 15:544 –549
A Comparison of Candesartan, Felodipine, and Their Combination in the Treatment of Elderly Patients With Systolic Hypertension Trefor Morgan and Adrianne Anderson Monotherapy frequently does not cause adequate blood pressure (BP) reduction and goal BP is not achieved. This double-blind, randomized, crossover, placebo-controlled study investigated, using a factorial design, the interaction between a dihydropyridine calcium channel blocking drug (felodipine 5 mg) and an angiotensin type 1 receptor blocking drug (candesartan 16 mg) on the control of BP as assessed by 24-h ambulatory monitoring. A total of 31 elderly patients with systolic hypertension completed all four arms of the study. Candesartan and felodipine lowered mean 24-h BP to a similar extent (candesartan 12.2 ⫾ 2.6/7.5 ⫾ 1.8; felodipine 11.9 ⫾ 2.2/5.7 ⫾ 1.4 mm Hg). The combination lowered it by 21.0 ⫾ 2.1/11.2 ⫾ 1.2 mm Hg, and this fall was significantly greater than with either of the monotherapies (P ⬍ .005) and was fully additive with no interactive term. The responder rate with the combination (90%) was greater than with candesartan (61%) or felo-
M
onotherapy achieves goal blood pressure (BP) in only a small percentage of elderly patients with predominant systolic hypertension. Thus, in many patients, two or more drugs are required.1,2 In a previous study3 we demonstrated that the use of felodipine (5 mg) together with enalapril (5 mg) provided better BP control than either drug alone in the same or a higher dose. The side effect profile was also improved, with a significant reduction in adverse events. These results were confirmed by others with the same4,5 or similar drug combinations,6,7 leading to the marketing of fixed-dose combinations of an angiotensin converting enzyme (ACE) inhibitor and a dihydropyridine calcium blocking drug. Angiotensin type 1 (AT1) receptor blocking drugs have more recently been introduced.8,9 These drugs have a similar antihypertensive effect to ACE inhibitors and are extremely well tolerated.10,11 Candesartan is a long-acting Received October 11, 2001. First Decision November 8, 2001. Accepted January 15, 2002. From the Department of Physiology, University of Melbourne and Hypertension Clinic, Austin and Repatriation Hospital Medical Center, Heidelberg, Australia. 0895-7061/02/$22.00 PII S0895-7061(02)02279-3
dipine (55%). Microalbuminuria or proteinuria was present in 12 of 31 patients at randomization despite previous BP control. Candesartan and the combination both reduced urinary albumin excretion. Albumin excretion was not reduced by felodipine despite BP control similar to that achieved with candesartan. Side effects were infrequent and were fewer on the combination than on placebo or on the monotherapies. The combination of felodipine 5 mg and candesartan 16 mg has additive effects on BP in elderly patients with systolic hypertension. The combination was well tolerated and is suitable for use in patients who do not have an adequate response to monotherapy. Am J Hypertens 2002;15:544 –549 © 2002 American Journal of Hypertension, Ltd. Key Words: Hypertension, elderly, systolic hypertension, microalbuminuria, ABPM, candesartan, felodipine, dihydropyridine calcium blockers, angiotensin receptor antagonists.
AT1-receptor blocker that has a slow dissociation from the receptor and provides BP control for ⱖ24 h.12 This study was designed to determine the interaction between candesartan (16 mg/day) and felodipine (5 mg/day) and whether the combination was more effective than either drug alone or the better of the two drugs. Effectiveness was determined by both clinic and ambulatory BP measurements.
Patients and Methods The patients all were ⬎65 years old and were known to have essential hypertension. All had previously been treated, and were known to have had a systolic blood pressure (SBP) ⱖ160 mm Hg before treatment. Patients were excluded from the study if they had cardiac failure or angina, a cerebral or cardiac event in the 6 months before the study or a creatinine ⬎0.16 mmol/L, liver enzymes more than three times normal, or any disease that was This study was supported by AstraZeneca. Address correspondence and reprint requests to Prof. T. Morgan, Department of Physiology, University of Melbourne, Parkville. Vic. 3010. Australia: e-mail: [email protected] © 2002 by the American Journal of Hypertension, Ltd. Published by Elsevier Science Inc.
AJH–June 2002–VOL. 15, NO. 6
CANDESARTAN, FELODIPINE, AND THEIR COMBINATION
likely to cause alterations of their regular medications during the study. The study, which was approved by the Ethics Committee of the Austin & Repatriation Medical Centre, was explained to the patients and informed consent obtained. Antihypertensive medication was stopped 3 to 7 days before their randomization visit, depending on the half-life of the drug in use. This was not to obtain a baseline BP but rather to exclude patients with an excessive rise in BP or symptoms with cessation of therapy. Patients provided a 24-h urine collection on which electrolytes, creatinine, proteinuria, and microalbuminuria were measured. Patients who met the eligibility criteria were randomized to a double-blind, four-way balanced-design crossover study in which they received placebo, candesartan cilexetil 16 mg, felodipine 5 mg, and candesartan cilexetil 16 mg ⫹ felodipine 5 mg. Each treatment period was 1 month. Patients took their medication between 7 AM and 9 AM. At each subsequent visit, patients attended the clinic between 7 AM and 9 AM and did not take their medication until after various measurements were made. Clinic BP was taken three times after 10 min supine and twice after 2 min standing, and the mean values recorded. An ambulatory BP monitor (A&D Co., Ltd., 2421, Tokyo, Japan) was then fitted to record BP three times/h using Korotkoff sounds and oscillometry for the next 25–26 h, and medication given. In all patients, medication was given before 9 AM. The patients returned to the clinic on the subsequent day at 9 AM without taking their medication. Clinic BP was recorded in the same way as stated above and the ambulatory BP monitor removed. Patients who had microalbuminuria (⬎8 g/min) or proteinuria at the randomization visit made a 24-h urine collection during each ambulatory BP measurement interval. If the ambulatory BP monitor did not meet satisfactory criteria, therapy was continued and the recording repeated within the next 5 days. Patients then progressed to the next treatment arm. When patients had completed the four-way crossover study they were given candesartan 32 mg/day as open medication. The ambulatory blood pressure monitor and clinic visits were repeated 4 to 6 weeks later. This continuation was only performed if a 24-h SBP ⱕ180 mm Hg had been recorded on one or more of the double-blind treatment periods. At each visit patients were asked if they had any side effects, and an assessment was made as to whether this was likely related to drug therapy. To be included in the statistical analysis patients were required to have a mean daytime SBP ⱖ135 mm Hg during the randomized placebo period. If the BP was below this value they were considered normotensive and excluded from efficacy analysis. The ambulatory BP measurements were also subdivided into daytime (time of application of ABPM until 10 PM), nighttime 12 midnight to 6 AM), and morning (6 AM to end of monitoring). The primary analysis compared the fall in 24-h SBP on combined therapy with the fall on candesartan or felodip-
545
ine and also with the greater of the falls on candesartan or felodipine. The fall in BP was defined as the difference between active and placebo treatment. Secondary analyses were related to the responses at different times of the day and to clinic BP. Effects on albumin excretion arising from the different drug regimens were compared and correlated with the BP response. Comparisons were made using analysis of variance and with paired t tests using the Bonferoni correction. In appropriate situations rank tests were used. Side effects were also described.
Results A total of 42 patients agreed to participate in the study. Two were excluded because of excessive BP rise or symptoms after previous medication was stopped; thus, 40 patients were randomized. After their medication was dispensed, three patients decided not to proceed with the study, took no medication, and had no subsequent visits. Before the first postrandomization visit one patient developed third degree heart block while on candesartan and was withdrawn. After the first visit a second patient withdrew because he believed that repeated ambulatory monitoring was too great an inconvenience. Two other patients withdrew within 1 week after visit 1 because of side effects (see later here). One patient was subsequently withdrawn because of difficulty following the dosing instructions, leading to haphazard compliance. One patient was not eligible for analysis because of a daytime SBP on placebo of 132 mm Hg. Thus, 31 patients completed the study with ambulatory BP recording on each therapeutic regimen. This is the data set analysed. All randomized patients who took any therapy (n ⫽ 37) were assessed for side effects. The demographic characteristics of the patients are indicated in Table 1. The 24-h ambulatory BP profiles during the different treatment periods are shown in Fig. 1. The 24-h SBP and diastolic BP (DBP) fell significantly with candesartan, felodipine and the combination (Fig. 2, Table 2). The fall in SBP with the combination was significantly greater than the fall with either candesartan or felodipine, and the fall with combined therapy was greater compared with the better of the response with candesartan or felodipine. The mean fall in 24-h SBP was 12.2 ⫾ 2.6 mm Hg with candesartan, 11.9 ⫾ 2.2 mm Hg with felodipine, and 21.0 ⫾ 2.1 mm Hg with the combination. The effect of the combination was fully additive, with no interactive term. The mean fall, taking the greater of the fall with candesartan or felodipine, was 4.7 ⫾ 1.6 mm Hg less than the fall with combination therapy (P ⫽ .005). The BP at various times of the day using the ambulatory BP monitor showed a similar response (Table 1). The fall in DBP with each of the regimens followed the same pattern as the SBP, although the falls were numerically smaller (Table 2). The mean falls in BP with candesartan and felodipine (both systolic and diastolic) did not differ, but different patients appeared to respond differently
546
CANDESARTAN, FELODIPINE, AND THEIR COMBINATION
AJH–June 2002–VOL. 15, NO. 6
Table 1. Demographics of patients who completed the study (n ⫽ 31) Age (y) Mean Range Height, cm Weight, Kg BMI M:F ratio BP on previous treatment Systolic Diastolic Previous ACEI Ca B Diuretics Other No drugs 1 Drug 2 Drugs ACEI ⫹ Ca B Other combination 3 Drugs 4 Drugs Diabetic Increased albumin excretion Diabetic with increased albumin excretion
77.0 ⫾ 0.8 65–86 166.6 ⫾ 1.3 79.1 ⫾ 2.4 28.4 ⫾ 0.7 28:3 152 ⫾ 2.6 80 ⫾ 1.4 22 18 11 6 0 12 9 4 5 1 4 12 2
BMI ⫽ body mass index; M:F ⫽ male:female; BP ⫽ blood pressure; ACEI ⫽ angiotensin converting enzyme inhibitor; Ca B ⫽ calcium blockers.
to one or the other drug. The response of the individual patients for 24-h SBP is indicated in Fig. 3. There were no significant changes in pulse rate (Table 2). The pulse pressures from the 24-h ambulatory BP monitor were placebo 79.4 ⫾ 2.6 mm Hg, candesartan 74.0 ⫾ 2.7 mm Hg, felodipine 72.8 ⫾ 2.3 mm Hg, and combination 68.5 ⫾ 2.5 mm Hg. All drugs caused a reduction in pulse pressure compared with placebo (P ⬍ .01), and the
FIG. 1. Systolic blood pressure at hourly intervals on placebo, felodipine (5 mg), candesartan (16 mg),or their combination. n ⫽ 31. Arrow indicates time of drug administration. Sys BP ⫽ systolic blood pressure.
FIG. 2. Fall in systolic blood pressure in patients treated with candesartan (Cand.) 16 mg, felodipine (Felod.) 5 mg, combination (Comb.), and the greater of the fall with candesartan or felodipine n ⫽ 31. Fall in blood pressure was the difference compared with the randomized placebo period. Other abbreviation as in Fig. 1.
fall with the combination was greater than with either monotherapy (P ⬍ .01). On placebo, the difference between daytime and nighttime BP was 18.5 ⫾ 3.0/9.2 ⫾ 1.6 mm Hg. The day–night differences on felodipine (16.9 ⫾ 3.1/10.0 ⫾ 1.3) and the combination (17.0 ⫾ 2.7/10.1 ⫾ 1.3) were similar to those on placebo. The difference between daytime and nighttime BP was significantly greater on candesartan monotherapy (22.5 ⫾ 2.4/12.6 ⫾ 1.1) than on placebo or the other treatments (P ⬍ .05). On felodipine or the combination the falls in nighttime and daytime BP were similar. However, in patients on candesartan, the fall in nighttime BP 14.6 ⫾ 3.0/8.9 ⫾ 1.5 was significantly greater (P ⬍ .01) than the fall in daytime BP 10.6 ⫾ 2.8/5.5 ⫾ 1.0 mm Hg and greater (P ⬍ .05) than the fall in nighttime BP on felodipine 10.1 ⫾ 2.6/5.5 ⫾ 1.6 mm Hg. This greater fall was achieved despite a lower SBP at night on placebo. Of the 31 patients, 22 achieved a 24-h SBP ⱕ150 mm Hg and a fall of ⱖ10 mm Hg on the combination, compared with 13 patients on candesartan and eight on felodipine (Table 3). The number of responders (SBP fall of ⱖ10 mm Hg) and the number who achieved goal (SBP ⱕ140 mm Hg) were greater in the combination group than in the other groups (Table 3). SBP and DBP measured at the clinic showed similar results as those demonstrated with ambulatory BP (Table 2). However, the falls in general were smaller, and not all of them reached significance. Twelve patients (39%) had increased protein excretion at the randomization visit and had their albumin excretion measured by 24-h urine collections on the day of the monitor at all subsequent visits. On placebo, the urinary albumin excretion rate was 192 g/min. The excretion rate of albumin was reduced in patients on candesartan (101 g/min) and in patients on combination therapy (93 g/ min). In patients on felodipine, despite a similar reduction
AJH–June 2002–VOL. 15, NO. 6
CANDESARTAN, FELODIPINE, AND THEIR COMBINATION
547
Table 2. Ambulatory and clinic blood pressure (mm Hg) and pulse (beats/min)
24-h Systolic Diastolic Pulse Daytime Systolic Diastolic Pulse Nighttime Systolic Diastolic Pulse Morning Systolic Diastolic Pulse Clinic Systolic Diastolic Pulse
Placebo
Candesartan
Felodipine
Combination
162.9 ⫾ 3.3 83.5 ⫾ 1.4 72.8 ⫾ 1.6
150.0 ⫾ 3.7 76.0 ⫾ 1.6 71.5 ⫾ 1.6
150.6 ⫾ 2.7 77.8 ⫾ 1.4 72.7 ⫾ 1.4
140.8 ⫾ 3.0* 72.3 ⫾ 1.3* 72.5 ⫾ 1.5
166.5 ⫾ 3.4 85.0 ⫾ 1.5 74.7 ⫾ 1.4
155.9 ⫾ 4.0 79.5 ⫾ 1.7 74.9 ⫾ 1.7
154.1 ⫾ 3.0 80.2 ⫾ 1.5 76.4 ⫾ 1.6
144.0 ⫾ 3.3* 74.2 ⫾ 1.4* 76.1 ⫾ 1.6
148.0 ⫾ 4.1 75.8 ⫾ 2.0 67.3 ⫾ 2.1
133.5 ⫾ 4.1 66.0 ⫾ 1.6 64.9 ⫾ 1.7
137.2 ⫾ 3.3 70.2 ⫾ 1.5 65.6 ⫾ 1.5
127.0 ⫾ 3.5† 64.1 ⫾ 1.5† 64.4 ⫾ 1.5
168.6 ⫾ 3.9 86.0 ⫾ 2.0 76.6 ⫾ 1.8
158.4 ⫾ 4.4 78.7 ⫾ 1.8 75.2 ⫾ 2.0
156.3 ⫾ 3.4 79.6 ⫾ 2.0 74.2 ⫾ 1.6
147.8 ⫾ 3.6* 75.5 ⫾ 1.8‡ 74.5 ⫾ 1.7
168.9 ⫾ 3.5 88.6 ⫾ 2.4 73.2 ⫾ 2.2
158.0 ⫾ 4.5 80.0 ⫾ 1.8 71.9 ⫾ 1.8
160.8 ⫾ 3.0 82.8 ⫾ 2.0 70.6 ⫾ 2.0
150.5 ⫾ 3.8§ 77.9 ⫾ 2.0 72.9 ⫾ 1.9
Mean ⫾ sem; n ⫽ 31. All systolic and diastolic blood pressure values on therapy are less than on placebo (P ⬍ .001 for ambulatory blood pressure monitoring and P ⬍ .05 for clinic). * P ⬍ .005 v candesartan and felodipine. † P ⬍ .001 v felodipine and ⬍ .05 v candesartan. ‡ P ⫽ .05 v candesartan and .026 v felodipine. § P ⬍ .05 v candesartan and felodipine.
in BP to that achieved with candesartan, the urine excretion of protein (226 g/min) was not different from that on placebo and was greater than the excretion rate on candesartan or on the combination. The changes in albumin excretion in individual patients are shown in Fig. 3. The relationship between the 24-h SBP and the albumin excretion rate is indicated in Fig. 4. For the same level of BP, patients on felodipine tended to have a higher urinary protein excretion compared with patients on candesartan or the combination. Three patients were withdrawn from the study because of adverse events. In two patients who had lethargy and pulmonary edema, this was in the first week of an active treatment period, which followed a placebo period and reflected poor control on placebo. Side effects occurred in seven patients on placebo, five on candesartan, four on felodipine, and two on the combination. Headache was more frequent on placebo than on other treatments. Ankle swelling was uncommon. At the end of the study, patients were asked if they wanted to continue in an open therapeutic extension in which they received candesartan 32 mg daily. Five patients declined to take part in this extension. One patient was excluded because SBP had not been reduced to ⬍180 mm Hg on any therapy and he was symptomatic. Thus, 25 patients were studied. The 24-h mean BP achieved with candesartan 32 mg of 150.6 ⫾ 4.1/79.0 ⫾ 1.8 mm Hg was higher than that achieved with combination therapy of 141.3 ⫾ 2.8/72.5 ⫾ 1.4 mm Hg (P ⬍ .001).
Discussion Candesartan and felodipine were both effective antihypertensive drugs, but the combination of the two caused a greater response that was fully additive with no interactive term. A greater response to combination therapy may be due to recruitment of nonresponders to one or the other drug. However, the response to the combination was greater than the better of the response to either drug, indicating that there was also a significant additive effect of the two drugs. The falls in BP were seen with the 24-h, daytime, nighttime, morning, and clinic BP. The fall in BP at nighttime with candesartan was greater than the daytime fall, indicating that BP at nighttime when patients are asleep is more dependent on angiotensin II than during the day. This supports the results seen with perindopril administered in the morning or the evening,13 and with the apparent disappearance of trandolapril effect between 30 and 38 h but the reappearance of the effect between 38 and 48 h when patients were asleep.14 The study indicated that ambulatory BP measurement is a more effective way of determining the response to a drug. The falls were greater with ambulatory BP monitoring, and statistical significance was achieved more consistently than with clinic BP measurements. Furthermore, there was a much greater variability in response with clinic BP than with ambulatory BP monitoring. In many paralleldesign studies with a run-in placebo period, the falls in clinic BP are usually greater than on ambulatory BP mon-
548
CANDESARTAN, FELODIPINE, AND THEIR COMBINATION
AJH–June 2002–VOL. 15, NO. 6
FIG. 4. Albumin excretion rate of individual patients on the different treatments plotted against the 24-hour systolic blood pressure (SBP). Abbreviations as in Figs. 2 and 3.
FIG. 3. Individual systolic blood pressure (24-h mean) and albumin excretion rate during the four periods of the study. Albumin excretion rate of individual patients on the three treatment regimens are expressed as a percentage of the excretion rate on placebo. Abbreviations as in Figs. 1 and 2.
itoring. This is almost certainly an effect of patient habituation, observer bias, and regression to the mean, which was avoided in this study by the randomized placebo period. Table 3. Number of patients who achieved systolic blood pressure below various values (mm Hg) on the different drug regimens <135 <140 <150 Placebo Candesartan Felodipine Combination
0 8 5 12
2 9 7 14*
8 17 17 22
Fall >10
<150 ⴙ Fall >10
19 17 28* (90%)
13 8 22*
n ⫽ 31. * Fishers exact test; P ⬍ .05. Combination better than other therapies.
Pulse pressure has recently been stated to be a risk factor for outcome, although it may be more correctly a risk predictor. The pulse pressure on all therapies was less than on placebo at clinic and on ambulatory BP monitoring, and the greatest fall was with the combination. Thus, if it is desirable to reduce pulse pressure, this was an effective regimen. The response shown in the add-on study indicated that adding a dihydropyridine calcium blocking drug (felodipine) to therapy was more effective than doubling the dose of an AT1 receptor blocking drug. A significant number of this hypertensive patient population (12/31, 39%) had increased albumin excretion at randomization despite previous treatment. Despite a reduction in BP, felodipine had no effect on albumin excretion. However, candesartan reduced urinary albumin excretion, and this reduction was maintained on combination therapy. Blood pressure lowering is probably important to reduce albumin excretion, but these results indicate that AT1 receptor blocking drugs have specific additional effects. It is possible that the greater reduction in albuminuria with candesartan compared with felodipine was due to the greater fall in BP at night. The three therapies were all well tolerated, and the side effects were probably mainly due to inadequate control of BP. Effects were relatively minor. In two of the three patients in whom therapy was stopped, this was within 1 week of the end of the placebo phase and probably reflected poor BP control at that time. The combination of felodipine and candesartan is an
AJH–June 2002–VOL. 15, NO. 6
CANDESARTAN, FELODIPINE, AND THEIR COMBINATION
effective and well tolerated antihypertensive therapy and is more effective than the individual components. In elderly persons with systolic hypertension, it lowered mean 24-h BP by 21/10 mm Hg compared with placebo and reduced proteinuria, with a minimal number of side effects. This combination is suitable for use in persons with systolic hypertension who do not achieve control with either drug as monotherapy.
6.
7.
8.
Acknowledgments The help of Denise Bertram and Jann Lauri in measuring BP is acknowledged, as is the help of Jan Lovett in typing the manuscript.
9.
References 1.
2.
3.
4.
5.
Hansson L, Zanchetti A, Carruthers SG, Dahlof B, Elmfeldt D, Julius S, Menard J, Rahn KH, Wedel H, Westerling C, for the HOT Study Group: Effects of intensive blood-pressure lowering and lowdose aspirin in patients with hypertension: principal results of the Hypertension Optimal Treatment (HOT) randomised trial. Lancet 1998;35:1755–1762. Morgan TO, Anderson AIE, MacInnis RJ: ACE inhibitors, betablockers, calcium blockers and diuretics for the control of systolic hypertension. Am J Hypertens 2001;14:241–247. Morgan TO, Anderson A, Jones E: Comparison and interaction of low dose felodipine (ER) and enalapril in the treatment of essential hypertension in elderly patients. Am J Hypertens 1992;5:238 –243. Gradman AH, Cutler NR, Davis PJ, Robbins JA, Weiss RJ, Wood BC, Michelson EL: Long-term efficacy, tolerability, and safety of the combination of enalapril and felodipine ER in the treatment of hypertension. Enalapril-Felodipine ER Factorial Study Group. Clin Ther 1998;20:527–538. Gradman AH, Cutler NR, Davis PH, Robbins JA, Weiss RJ, Wood BC: Combined enalapril and felodipine extended release (ER) for
10.
11.
12.
13.
14.
549
systemic hypertension. Enalapril-Felodipine ER Factorial Study Group. Am J Cardiol 1997;79:431–435. Messerli F, Frishman WH, Elliott WJ: Effects of verapamil and trandolapril in the treatment of hypertension. Trandolapril Study Group. Am J Hypertens 1998;11(1);:322–327. Kuschnir E, Acuna E, Sevilla D: Treatment of patients with essential hypertension: amlodipine 5 mg/benazepril 20 mg compared with amlodipine 5 mg, benazepril 20 mg, and placebo. Clin Ther 1996; 18:1213–1223. Gradman AH, Arcuri KE, Goldberg AL, Ikeda LS, Nelson EB, Snavely DB, Sweet CS: A randomized, placebo-controlled, doubleblind, parallel study of various doses of losartan potassium compared with enalapril maleate in patients with essential hypertension. Hypertension 1995;25:1345–1350. Lacourcie`re Y, Asmar R, for the Candesartan/Losartan Study Investigators: A comparison of the efficacy and duration of action of candesartan cilexetil and losartan as assessed by clinic and ambulatory blood pressure after a missed dose, in truly hypertensive patients: A placebo-controlled, forced titration study. Am J Hypertens 1999;12:1181–1187. Lacourcie`re Y, Brunner H, Irwin R, Karlberg BE, Ramsey LE, Snavely DB, Dobbins TW, Faison EP, Nelson EB, and the Losartan Cough Study Group: Effects of modulators of the renin-angiotensinaldosterone system on cough. J Hypertens 1994;12:1387–1393. Sever P, Holzgreve H: Long term efficacy and tolerability of candesartan cilexatil in patients with mild to moderation hypertension. J Hum Hypertens 1997;11(Suppl 2):S69 –S73. Morsing P, Adler G, Brandt-Eliason U, Karp L, Ohlson K, Renberg L, Sojquist PO, Abrahamsson T: Mechanistic differences of various AT1-receptor blockers in isolated vessels of different origin. Hypertension 1999;33:1406 –1413. Morgan T, Anderson A, Jones E: The effect of 24h blood pressure control of an angiotensin converting enzyme inhibitor (perindopril) administered in the morning or at night. J Hypertens 1997;15:205– 211. Smith DHG, Neutel JM, Black HR, Schoenberger JA, Weber MA: Once-daily therapy with trandolapril in the treatment of hypertension. J Hum Hypertens 1996;10:129 –134.
2002; 15:544 –549
A Comparison of Candesartan, Felodipine, and Their Combination in the Treatment of Elderly Patients With Systolic Hypertension Trefor Morgan and Adrianne Anderson Monotherapy frequently does not cause adequate blood pressure (BP) reduction and goal BP is not achieved. This double-blind, randomized, crossover, placebo-controlled study investigated, using a factorial design, the interaction between a dihydropyridine calcium channel blocking drug (felodipine 5 mg) and an angiotensin type 1 receptor blocking drug (candesartan 16 mg) on the control of BP as assessed by 24-h ambulatory monitoring. A total of 31 elderly patients with systolic hypertension completed all four arms of the study. Candesartan and felodipine lowered mean 24-h BP to a similar extent (candesartan 12.2 ⫾ 2.6/7.5 ⫾ 1.8; felodipine 11.9 ⫾ 2.2/5.7 ⫾ 1.4 mm Hg). The combination lowered it by 21.0 ⫾ 2.1/11.2 ⫾ 1.2 mm Hg, and this fall was significantly greater than with either of the monotherapies (P ⬍ .005) and was fully additive with no interactive term. The responder rate with the combination (90%) was greater than with candesartan (61%) or felo-
M
onotherapy achieves goal blood pressure (BP) in only a small percentage of elderly patients with predominant systolic hypertension. Thus, in many patients, two or more drugs are required.1,2 In a previous study3 we demonstrated that the use of felodipine (5 mg) together with enalapril (5 mg) provided better BP control than either drug alone in the same or a higher dose. The side effect profile was also improved, with a significant reduction in adverse events. These results were confirmed by others with the same4,5 or similar drug combinations,6,7 leading to the marketing of fixed-dose combinations of an angiotensin converting enzyme (ACE) inhibitor and a dihydropyridine calcium blocking drug. Angiotensin type 1 (AT1) receptor blocking drugs have more recently been introduced.8,9 These drugs have a similar antihypertensive effect to ACE inhibitors and are extremely well tolerated.10,11 Candesartan is a long-acting Received October 11, 2001. First Decision November 8, 2001. Accepted January 15, 2002. From the Department of Physiology, University of Melbourne and Hypertension Clinic, Austin and Repatriation Hospital Medical Center, Heidelberg, Australia. 0895-7061/02/$22.00 PII S0895-7061(02)02279-3
dipine (55%). Microalbuminuria or proteinuria was present in 12 of 31 patients at randomization despite previous BP control. Candesartan and the combination both reduced urinary albumin excretion. Albumin excretion was not reduced by felodipine despite BP control similar to that achieved with candesartan. Side effects were infrequent and were fewer on the combination than on placebo or on the monotherapies. The combination of felodipine 5 mg and candesartan 16 mg has additive effects on BP in elderly patients with systolic hypertension. The combination was well tolerated and is suitable for use in patients who do not have an adequate response to monotherapy. Am J Hypertens 2002;15:544 –549 © 2002 American Journal of Hypertension, Ltd. Key Words: Hypertension, elderly, systolic hypertension, microalbuminuria, ABPM, candesartan, felodipine, dihydropyridine calcium blockers, angiotensin receptor antagonists.
AT1-receptor blocker that has a slow dissociation from the receptor and provides BP control for ⱖ24 h.12 This study was designed to determine the interaction between candesartan (16 mg/day) and felodipine (5 mg/day) and whether the combination was more effective than either drug alone or the better of the two drugs. Effectiveness was determined by both clinic and ambulatory BP measurements.
Patients and Methods The patients all were ⬎65 years old and were known to have essential hypertension. All had previously been treated, and were known to have had a systolic blood pressure (SBP) ⱖ160 mm Hg before treatment. Patients were excluded from the study if they had cardiac failure or angina, a cerebral or cardiac event in the 6 months before the study or a creatinine ⬎0.16 mmol/L, liver enzymes more than three times normal, or any disease that was This study was supported by AstraZeneca. Address correspondence and reprint requests to Prof. T. Morgan, Department of Physiology, University of Melbourne, Parkville. Vic. 3010. Australia: e-mail: [email protected] © 2002 by the American Journal of Hypertension, Ltd. Published by Elsevier Science Inc.
AJH–June 2002–VOL. 15, NO. 6
CANDESARTAN, FELODIPINE, AND THEIR COMBINATION
likely to cause alterations of their regular medications during the study. The study, which was approved by the Ethics Committee of the Austin & Repatriation Medical Centre, was explained to the patients and informed consent obtained. Antihypertensive medication was stopped 3 to 7 days before their randomization visit, depending on the half-life of the drug in use. This was not to obtain a baseline BP but rather to exclude patients with an excessive rise in BP or symptoms with cessation of therapy. Patients provided a 24-h urine collection on which electrolytes, creatinine, proteinuria, and microalbuminuria were measured. Patients who met the eligibility criteria were randomized to a double-blind, four-way balanced-design crossover study in which they received placebo, candesartan cilexetil 16 mg, felodipine 5 mg, and candesartan cilexetil 16 mg ⫹ felodipine 5 mg. Each treatment period was 1 month. Patients took their medication between 7 AM and 9 AM. At each subsequent visit, patients attended the clinic between 7 AM and 9 AM and did not take their medication until after various measurements were made. Clinic BP was taken three times after 10 min supine and twice after 2 min standing, and the mean values recorded. An ambulatory BP monitor (A&D Co., Ltd., 2421, Tokyo, Japan) was then fitted to record BP three times/h using Korotkoff sounds and oscillometry for the next 25–26 h, and medication given. In all patients, medication was given before 9 AM. The patients returned to the clinic on the subsequent day at 9 AM without taking their medication. Clinic BP was recorded in the same way as stated above and the ambulatory BP monitor removed. Patients who had microalbuminuria (⬎8 g/min) or proteinuria at the randomization visit made a 24-h urine collection during each ambulatory BP measurement interval. If the ambulatory BP monitor did not meet satisfactory criteria, therapy was continued and the recording repeated within the next 5 days. Patients then progressed to the next treatment arm. When patients had completed the four-way crossover study they were given candesartan 32 mg/day as open medication. The ambulatory blood pressure monitor and clinic visits were repeated 4 to 6 weeks later. This continuation was only performed if a 24-h SBP ⱕ180 mm Hg had been recorded on one or more of the double-blind treatment periods. At each visit patients were asked if they had any side effects, and an assessment was made as to whether this was likely related to drug therapy. To be included in the statistical analysis patients were required to have a mean daytime SBP ⱖ135 mm Hg during the randomized placebo period. If the BP was below this value they were considered normotensive and excluded from efficacy analysis. The ambulatory BP measurements were also subdivided into daytime (time of application of ABPM until 10 PM), nighttime 12 midnight to 6 AM), and morning (6 AM to end of monitoring). The primary analysis compared the fall in 24-h SBP on combined therapy with the fall on candesartan or felodip-
545
ine and also with the greater of the falls on candesartan or felodipine. The fall in BP was defined as the difference between active and placebo treatment. Secondary analyses were related to the responses at different times of the day and to clinic BP. Effects on albumin excretion arising from the different drug regimens were compared and correlated with the BP response. Comparisons were made using analysis of variance and with paired t tests using the Bonferoni correction. In appropriate situations rank tests were used. Side effects were also described.
Results A total of 42 patients agreed to participate in the study. Two were excluded because of excessive BP rise or symptoms after previous medication was stopped; thus, 40 patients were randomized. After their medication was dispensed, three patients decided not to proceed with the study, took no medication, and had no subsequent visits. Before the first postrandomization visit one patient developed third degree heart block while on candesartan and was withdrawn. After the first visit a second patient withdrew because he believed that repeated ambulatory monitoring was too great an inconvenience. Two other patients withdrew within 1 week after visit 1 because of side effects (see later here). One patient was subsequently withdrawn because of difficulty following the dosing instructions, leading to haphazard compliance. One patient was not eligible for analysis because of a daytime SBP on placebo of 132 mm Hg. Thus, 31 patients completed the study with ambulatory BP recording on each therapeutic regimen. This is the data set analysed. All randomized patients who took any therapy (n ⫽ 37) were assessed for side effects. The demographic characteristics of the patients are indicated in Table 1. The 24-h ambulatory BP profiles during the different treatment periods are shown in Fig. 1. The 24-h SBP and diastolic BP (DBP) fell significantly with candesartan, felodipine and the combination (Fig. 2, Table 2). The fall in SBP with the combination was significantly greater than the fall with either candesartan or felodipine, and the fall with combined therapy was greater compared with the better of the response with candesartan or felodipine. The mean fall in 24-h SBP was 12.2 ⫾ 2.6 mm Hg with candesartan, 11.9 ⫾ 2.2 mm Hg with felodipine, and 21.0 ⫾ 2.1 mm Hg with the combination. The effect of the combination was fully additive, with no interactive term. The mean fall, taking the greater of the fall with candesartan or felodipine, was 4.7 ⫾ 1.6 mm Hg less than the fall with combination therapy (P ⫽ .005). The BP at various times of the day using the ambulatory BP monitor showed a similar response (Table 1). The fall in DBP with each of the regimens followed the same pattern as the SBP, although the falls were numerically smaller (Table 2). The mean falls in BP with candesartan and felodipine (both systolic and diastolic) did not differ, but different patients appeared to respond differently
546
CANDESARTAN, FELODIPINE, AND THEIR COMBINATION
AJH–June 2002–VOL. 15, NO. 6
Table 1. Demographics of patients who completed the study (n ⫽ 31) Age (y) Mean Range Height, cm Weight, Kg BMI M:F ratio BP on previous treatment Systolic Diastolic Previous ACEI Ca B Diuretics Other No drugs 1 Drug 2 Drugs ACEI ⫹ Ca B Other combination 3 Drugs 4 Drugs Diabetic Increased albumin excretion Diabetic with increased albumin excretion
77.0 ⫾ 0.8 65–86 166.6 ⫾ 1.3 79.1 ⫾ 2.4 28.4 ⫾ 0.7 28:3 152 ⫾ 2.6 80 ⫾ 1.4 22 18 11 6 0 12 9 4 5 1 4 12 2
BMI ⫽ body mass index; M:F ⫽ male:female; BP ⫽ blood pressure; ACEI ⫽ angiotensin converting enzyme inhibitor; Ca B ⫽ calcium blockers.
to one or the other drug. The response of the individual patients for 24-h SBP is indicated in Fig. 3. There were no significant changes in pulse rate (Table 2). The pulse pressures from the 24-h ambulatory BP monitor were placebo 79.4 ⫾ 2.6 mm Hg, candesartan 74.0 ⫾ 2.7 mm Hg, felodipine 72.8 ⫾ 2.3 mm Hg, and combination 68.5 ⫾ 2.5 mm Hg. All drugs caused a reduction in pulse pressure compared with placebo (P ⬍ .01), and the
FIG. 1. Systolic blood pressure at hourly intervals on placebo, felodipine (5 mg), candesartan (16 mg),or their combination. n ⫽ 31. Arrow indicates time of drug administration. Sys BP ⫽ systolic blood pressure.
FIG. 2. Fall in systolic blood pressure in patients treated with candesartan (Cand.) 16 mg, felodipine (Felod.) 5 mg, combination (Comb.), and the greater of the fall with candesartan or felodipine n ⫽ 31. Fall in blood pressure was the difference compared with the randomized placebo period. Other abbreviation as in Fig. 1.
fall with the combination was greater than with either monotherapy (P ⬍ .01). On placebo, the difference between daytime and nighttime BP was 18.5 ⫾ 3.0/9.2 ⫾ 1.6 mm Hg. The day–night differences on felodipine (16.9 ⫾ 3.1/10.0 ⫾ 1.3) and the combination (17.0 ⫾ 2.7/10.1 ⫾ 1.3) were similar to those on placebo. The difference between daytime and nighttime BP was significantly greater on candesartan monotherapy (22.5 ⫾ 2.4/12.6 ⫾ 1.1) than on placebo or the other treatments (P ⬍ .05). On felodipine or the combination the falls in nighttime and daytime BP were similar. However, in patients on candesartan, the fall in nighttime BP 14.6 ⫾ 3.0/8.9 ⫾ 1.5 was significantly greater (P ⬍ .01) than the fall in daytime BP 10.6 ⫾ 2.8/5.5 ⫾ 1.0 mm Hg and greater (P ⬍ .05) than the fall in nighttime BP on felodipine 10.1 ⫾ 2.6/5.5 ⫾ 1.6 mm Hg. This greater fall was achieved despite a lower SBP at night on placebo. Of the 31 patients, 22 achieved a 24-h SBP ⱕ150 mm Hg and a fall of ⱖ10 mm Hg on the combination, compared with 13 patients on candesartan and eight on felodipine (Table 3). The number of responders (SBP fall of ⱖ10 mm Hg) and the number who achieved goal (SBP ⱕ140 mm Hg) were greater in the combination group than in the other groups (Table 3). SBP and DBP measured at the clinic showed similar results as those demonstrated with ambulatory BP (Table 2). However, the falls in general were smaller, and not all of them reached significance. Twelve patients (39%) had increased protein excretion at the randomization visit and had their albumin excretion measured by 24-h urine collections on the day of the monitor at all subsequent visits. On placebo, the urinary albumin excretion rate was 192 g/min. The excretion rate of albumin was reduced in patients on candesartan (101 g/min) and in patients on combination therapy (93 g/ min). In patients on felodipine, despite a similar reduction
AJH–June 2002–VOL. 15, NO. 6
CANDESARTAN, FELODIPINE, AND THEIR COMBINATION
547
Table 2. Ambulatory and clinic blood pressure (mm Hg) and pulse (beats/min)
24-h Systolic Diastolic Pulse Daytime Systolic Diastolic Pulse Nighttime Systolic Diastolic Pulse Morning Systolic Diastolic Pulse Clinic Systolic Diastolic Pulse
Placebo
Candesartan
Felodipine
Combination
162.9 ⫾ 3.3 83.5 ⫾ 1.4 72.8 ⫾ 1.6
150.0 ⫾ 3.7 76.0 ⫾ 1.6 71.5 ⫾ 1.6
150.6 ⫾ 2.7 77.8 ⫾ 1.4 72.7 ⫾ 1.4
140.8 ⫾ 3.0* 72.3 ⫾ 1.3* 72.5 ⫾ 1.5
166.5 ⫾ 3.4 85.0 ⫾ 1.5 74.7 ⫾ 1.4
155.9 ⫾ 4.0 79.5 ⫾ 1.7 74.9 ⫾ 1.7
154.1 ⫾ 3.0 80.2 ⫾ 1.5 76.4 ⫾ 1.6
144.0 ⫾ 3.3* 74.2 ⫾ 1.4* 76.1 ⫾ 1.6
148.0 ⫾ 4.1 75.8 ⫾ 2.0 67.3 ⫾ 2.1
133.5 ⫾ 4.1 66.0 ⫾ 1.6 64.9 ⫾ 1.7
137.2 ⫾ 3.3 70.2 ⫾ 1.5 65.6 ⫾ 1.5
127.0 ⫾ 3.5† 64.1 ⫾ 1.5† 64.4 ⫾ 1.5
168.6 ⫾ 3.9 86.0 ⫾ 2.0 76.6 ⫾ 1.8
158.4 ⫾ 4.4 78.7 ⫾ 1.8 75.2 ⫾ 2.0
156.3 ⫾ 3.4 79.6 ⫾ 2.0 74.2 ⫾ 1.6
147.8 ⫾ 3.6* 75.5 ⫾ 1.8‡ 74.5 ⫾ 1.7
168.9 ⫾ 3.5 88.6 ⫾ 2.4 73.2 ⫾ 2.2
158.0 ⫾ 4.5 80.0 ⫾ 1.8 71.9 ⫾ 1.8
160.8 ⫾ 3.0 82.8 ⫾ 2.0 70.6 ⫾ 2.0
150.5 ⫾ 3.8§ 77.9 ⫾ 2.0 72.9 ⫾ 1.9
Mean ⫾ sem; n ⫽ 31. All systolic and diastolic blood pressure values on therapy are less than on placebo (P ⬍ .001 for ambulatory blood pressure monitoring and P ⬍ .05 for clinic). * P ⬍ .005 v candesartan and felodipine. † P ⬍ .001 v felodipine and ⬍ .05 v candesartan. ‡ P ⫽ .05 v candesartan and .026 v felodipine. § P ⬍ .05 v candesartan and felodipine.
in BP to that achieved with candesartan, the urine excretion of protein (226 g/min) was not different from that on placebo and was greater than the excretion rate on candesartan or on the combination. The changes in albumin excretion in individual patients are shown in Fig. 3. The relationship between the 24-h SBP and the albumin excretion rate is indicated in Fig. 4. For the same level of BP, patients on felodipine tended to have a higher urinary protein excretion compared with patients on candesartan or the combination. Three patients were withdrawn from the study because of adverse events. In two patients who had lethargy and pulmonary edema, this was in the first week of an active treatment period, which followed a placebo period and reflected poor control on placebo. Side effects occurred in seven patients on placebo, five on candesartan, four on felodipine, and two on the combination. Headache was more frequent on placebo than on other treatments. Ankle swelling was uncommon. At the end of the study, patients were asked if they wanted to continue in an open therapeutic extension in which they received candesartan 32 mg daily. Five patients declined to take part in this extension. One patient was excluded because SBP had not been reduced to ⬍180 mm Hg on any therapy and he was symptomatic. Thus, 25 patients were studied. The 24-h mean BP achieved with candesartan 32 mg of 150.6 ⫾ 4.1/79.0 ⫾ 1.8 mm Hg was higher than that achieved with combination therapy of 141.3 ⫾ 2.8/72.5 ⫾ 1.4 mm Hg (P ⬍ .001).
Discussion Candesartan and felodipine were both effective antihypertensive drugs, but the combination of the two caused a greater response that was fully additive with no interactive term. A greater response to combination therapy may be due to recruitment of nonresponders to one or the other drug. However, the response to the combination was greater than the better of the response to either drug, indicating that there was also a significant additive effect of the two drugs. The falls in BP were seen with the 24-h, daytime, nighttime, morning, and clinic BP. The fall in BP at nighttime with candesartan was greater than the daytime fall, indicating that BP at nighttime when patients are asleep is more dependent on angiotensin II than during the day. This supports the results seen with perindopril administered in the morning or the evening,13 and with the apparent disappearance of trandolapril effect between 30 and 38 h but the reappearance of the effect between 38 and 48 h when patients were asleep.14 The study indicated that ambulatory BP measurement is a more effective way of determining the response to a drug. The falls were greater with ambulatory BP monitoring, and statistical significance was achieved more consistently than with clinic BP measurements. Furthermore, there was a much greater variability in response with clinic BP than with ambulatory BP monitoring. In many paralleldesign studies with a run-in placebo period, the falls in clinic BP are usually greater than on ambulatory BP mon-
548
CANDESARTAN, FELODIPINE, AND THEIR COMBINATION
AJH–June 2002–VOL. 15, NO. 6
FIG. 4. Albumin excretion rate of individual patients on the different treatments plotted against the 24-hour systolic blood pressure (SBP). Abbreviations as in Figs. 2 and 3.
FIG. 3. Individual systolic blood pressure (24-h mean) and albumin excretion rate during the four periods of the study. Albumin excretion rate of individual patients on the three treatment regimens are expressed as a percentage of the excretion rate on placebo. Abbreviations as in Figs. 1 and 2.
itoring. This is almost certainly an effect of patient habituation, observer bias, and regression to the mean, which was avoided in this study by the randomized placebo period. Table 3. Number of patients who achieved systolic blood pressure below various values (mm Hg) on the different drug regimens <135 <140 <150 Placebo Candesartan Felodipine Combination
0 8 5 12
2 9 7 14*
8 17 17 22
Fall >10
<150 ⴙ Fall >10
19 17 28* (90%)
13 8 22*
n ⫽ 31. * Fishers exact test; P ⬍ .05. Combination better than other therapies.
Pulse pressure has recently been stated to be a risk factor for outcome, although it may be more correctly a risk predictor. The pulse pressure on all therapies was less than on placebo at clinic and on ambulatory BP monitoring, and the greatest fall was with the combination. Thus, if it is desirable to reduce pulse pressure, this was an effective regimen. The response shown in the add-on study indicated that adding a dihydropyridine calcium blocking drug (felodipine) to therapy was more effective than doubling the dose of an AT1 receptor blocking drug. A significant number of this hypertensive patient population (12/31, 39%) had increased albumin excretion at randomization despite previous treatment. Despite a reduction in BP, felodipine had no effect on albumin excretion. However, candesartan reduced urinary albumin excretion, and this reduction was maintained on combination therapy. Blood pressure lowering is probably important to reduce albumin excretion, but these results indicate that AT1 receptor blocking drugs have specific additional effects. It is possible that the greater reduction in albuminuria with candesartan compared with felodipine was due to the greater fall in BP at night. The three therapies were all well tolerated, and the side effects were probably mainly due to inadequate control of BP. Effects were relatively minor. In two of the three patients in whom therapy was stopped, this was within 1 week of the end of the placebo phase and probably reflected poor BP control at that time. The combination of felodipine and candesartan is an
AJH–June 2002–VOL. 15, NO. 6
CANDESARTAN, FELODIPINE, AND THEIR COMBINATION
effective and well tolerated antihypertensive therapy and is more effective than the individual components. In elderly persons with systolic hypertension, it lowered mean 24-h BP by 21/10 mm Hg compared with placebo and reduced proteinuria, with a minimal number of side effects. This combination is suitable for use in persons with systolic hypertension who do not achieve control with either drug as monotherapy.
6.
7.
8.
Acknowledgments The help of Denise Bertram and Jann Lauri in measuring BP is acknowledged, as is the help of Jan Lovett in typing the manuscript.
9.
References 1.
2.
3.
4.
5.
Hansson L, Zanchetti A, Carruthers SG, Dahlof B, Elmfeldt D, Julius S, Menard J, Rahn KH, Wedel H, Westerling C, for the HOT Study Group: Effects of intensive blood-pressure lowering and lowdose aspirin in patients with hypertension: principal results of the Hypertension Optimal Treatment (HOT) randomised trial. Lancet 1998;35:1755–1762. Morgan TO, Anderson AIE, MacInnis RJ: ACE inhibitors, betablockers, calcium blockers and diuretics for the control of systolic hypertension. Am J Hypertens 2001;14:241–247. Morgan TO, Anderson A, Jones E: Comparison and interaction of low dose felodipine (ER) and enalapril in the treatment of essential hypertension in elderly patients. Am J Hypertens 1992;5:238 –243. Gradman AH, Cutler NR, Davis PJ, Robbins JA, Weiss RJ, Wood BC, Michelson EL: Long-term efficacy, tolerability, and safety of the combination of enalapril and felodipine ER in the treatment of hypertension. Enalapril-Felodipine ER Factorial Study Group. Clin Ther 1998;20:527–538. Gradman AH, Cutler NR, Davis PH, Robbins JA, Weiss RJ, Wood BC: Combined enalapril and felodipine extended release (ER) for
10.
11.
12.
13.
14.
549
systemic hypertension. Enalapril-Felodipine ER Factorial Study Group. Am J Cardiol 1997;79:431–435. Messerli F, Frishman WH, Elliott WJ: Effects of verapamil and trandolapril in the treatment of hypertension. Trandolapril Study Group. Am J Hypertens 1998;11(1);:322–327. Kuschnir E, Acuna E, Sevilla D: Treatment of patients with essential hypertension: amlodipine 5 mg/benazepril 20 mg compared with amlodipine 5 mg, benazepril 20 mg, and placebo. Clin Ther 1996; 18:1213–1223. Gradman AH, Arcuri KE, Goldberg AL, Ikeda LS, Nelson EB, Snavely DB, Sweet CS: A randomized, placebo-controlled, doubleblind, parallel study of various doses of losartan potassium compared with enalapril maleate in patients with essential hypertension. Hypertension 1995;25:1345–1350. Lacourcie`re Y, Asmar R, for the Candesartan/Losartan Study Investigators: A comparison of the efficacy and duration of action of candesartan cilexetil and losartan as assessed by clinic and ambulatory blood pressure after a missed dose, in truly hypertensive patients: A placebo-controlled, forced titration study. Am J Hypertens 1999;12:1181–1187. Lacourcie`re Y, Brunner H, Irwin R, Karlberg BE, Ramsey LE, Snavely DB, Dobbins TW, Faison EP, Nelson EB, and the Losartan Cough Study Group: Effects of modulators of the renin-angiotensinaldosterone system on cough. J Hypertens 1994;12:1387–1393. Sever P, Holzgreve H: Long term efficacy and tolerability of candesartan cilexatil in patients with mild to moderation hypertension. J Hum Hypertens 1997;11(Suppl 2):S69 –S73. Morsing P, Adler G, Brandt-Eliason U, Karp L, Ohlson K, Renberg L, Sojquist PO, Abrahamsson T: Mechanistic differences of various AT1-receptor blockers in isolated vessels of different origin. Hypertension 1999;33:1406 –1413. Morgan T, Anderson A, Jones E: The effect of 24h blood pressure control of an angiotensin converting enzyme inhibitor (perindopril) administered in the morning or at night. J Hypertens 1997;15:205– 211. Smith DHG, Neutel JM, Black HR, Schoenberger JA, Weber MA: Once-daily therapy with trandolapril in the treatment of hypertension. J Hum Hypertens 1996;10:129 –134.