- Email: [email protected]
Effect of Amlodipine Versus Felodipine Extended Release on 24-Hour Ambulatory Blood Pressure in Hypertension
AJH
1998;11:690 – 696
Effect of Amlodipine Versus Felodipine Extended Release on 24-Hour Ambulatory Blood Pressure in Hypertension ¨ stergren, Hans Isaksson, Ulf Brodin, Åke Schwan, and K. Peter O ¨ hman for the study group Jan O
Amlodipine and felodipine are calcium antagonists of the dihydropyridine type. The elimination halflife of amlodipine is longer than that of felodipine. To study whether the different elimination rates of the drugs were reflected in different duration of blood pressure (BP) control, we compared amlodipine and felodipine extended release (ER) by both conventional clinic BP 24 h after drug intake and 24 h ambulatory BP monitoring (ABPM), with special reference to nighttime and morning blood pressure. Two hundred and sixteen patients with primary hypertension (supine diastolic BP, 95 to 115 mm Hg) were randomized to receive amlodipine or felodipine ER in a multicenter study. The starting dose of both drugs was 5 mg. If the target clinic diastolic BP (90 mm Hg) had not been achieved after 4 weeks the dose was increased to 10 mg. Twenty-four-hour ABPM was performed with the subjects taking placebo medication before randomization and after 4 and 8 weeks undergoing active treatment.
B
Significantly more patients responded after 4 weeks of treatment with amlodipine (50%) as compared with felodipine (33%) (P 5 .013). ABPM during daytime (07:00 to 23:00) was similar during both treatments, but nighttime systolic (P 5 .026) and diastolic (P 5 .019) BP was more effectively reduced by amlodipine than by felodipine. After 8 weeks 82% achieved the target pressure with amlodipine and 69% with felodipine (P 5 .036 for the difference). Amlodipine seems to be more effective than felodipine when the drugs are compared in the same dose, with regard to the effect on clinic BP 24 h after dosing and to ambulatory BP during the night. The longer elimination half-life of amlodipine as compared to felodipine is the probable reason for this finding. Am J Hypertens 1998;11:690 – 696 © 1998 American Journal of Hypertension, Ltd. KEY WORDS:
Ambulatory blood pressure, circadian variation, amlodipine, felodipine, plasma half-life.
lood pressure (BP) increases rather rapidly during the early morning hours1,2 and in some cases the increase in BP may contribute to the increase in cardiovascular events seen during this time period.3,4 Therefore, it may be impor-
tant to have an adequate BP-lowering effect during the entire 24-h period to avoid pressure peaks at the end of the dose interval when treating hypertensive subjects with a drug that is given once daily.5 The dihydropyridine calcium antagonists amlodip-
Received April 23, 1997. Accepted January 20, 1998. ¨ ); DeFrom the Department of Medicine, Karolinska Hospital (JO partment of Medicine, So¨dersjukhuset (HI); Medical Statistics Unit, Karolinska Institute (UB), Stockholm; Department of Family Medicine, Uppsala University (AS); and the Department of Endocrinol¨ ), Linko¨ping, Sweden. ogy, University Hospital (KPO
This study was supported by Pfizer AB, Sweden. ¨ sterAddress for correspondence and reprint requests to Jan O gren, MD, PhD, Department of Medicine, Karolinska Hospital, 171 76 Stockholm, Sweden; e-mail: [email protected]
© 1998 by the American Journal of Hypertension, Ltd. Published by Elsevier Science, Inc.
0895-7061/98/$19.00 PII S0895-7061(98)00032-6
AJH–JUNE 1998 –VOL. 11, NO. 6, PART 1
24-HOUR BLOOD PRESSURE WITH AMLODIPINE V FELODIPINE
ine and felodipine are frequently used in the treatment of hypertension. Peak plasma level for amlodipine occurs 6 to 12 h after oral tablet intake, with an elimination half-life of 35 to 50 h.6 For felodipine, the peak plasma concentration occurs after 1 to 3 h and the elimination half-life is 6 to 8 h.7 With felodipine as an extended release (ER) formulation, the time to peak plasma concentration is prolonged to 3 to 7 h after dosing and the plasma concentration at 24 h after dosing is increased.7,8 The primary objective of this study was to investigate whether starting doses of 5 mg of amlodipine versus 5 mg of felodipine ER are equipotent in the treatment of primary hypertension with focus on the antihypertensive effect at the end of the dosing interval. Twenty-four-hour ambulatory BP monitoring (ABPM) was used to allow analysis of the BP-lowering effect of the drugs in the later part of the dosing interval (ie, during the night and the early morning hours). The primary efficacy variable was the decrease in mean diastolic BP in the last 4 h in the 24-h ABPMs after 4 weeks of treatment with 5 mg of either drug. Secondary objectives were 1) to investigate the effect on 24-h ABPM of uptitrated doses to 10 mg when necessary, 2) to investigate the effect on clinic BP, 3) to investigate subjective side effects and ankle edema, and 4) to investigate the effect of treatment on fasting plasma lipids. In addition, plasma concentrations of the drugs were measured in conjunction with clinic BP at the end of the dose interval. PATIENTS AND METHODS The study was a single-blind, multicenter, randomized, parallel group study. Patients with mild-to-moderate primary hypertension (mean supine diastolic BP, 95 to 115 mm Hg) were enrolled at 13 centers in Sweden by 18 investigators. Patients were excluded if they had insulin-treated diabetes mellitus, renal impairment, or if they had experienced a cerebrovascular accident within the past 3 months and angina or coronary infarction within the past 6 months. Medication that might alter BP and changes in lipid-lowering treatment were not allowed in the study. Patients who had previously been treated with antihypertensive drugs had their medication withdrawn before entering the study. The study was initially planned for 140 patients to complete the study. One hundred and sixty patients were initially enrolled. A preplanned interim statistical analysis of the variation in BP changes during the last 4 h of the recording was performed to investigate the power of the study. It was then decided that an additional 56 patients were to be enrolled. Thus, 216 patients (120 men and 96 women) were entered into the study and randomized to treatment with amlodip-
691
TABLE 1. BASELINE DATA FOR ALL PATIENTS, FROM VISIT WEEK 0 Factor
Amlodipine
Felodipine
Age (years) Gender (percent men) Weight (kg) Systolic BP, supine (mm Hg) Diastolic BP, supine (mm Hg) Heart rate, supine Previous treatment for hypertension (%) Smokers (no.) Fasting glucose (mmol/L) Total cholesterol (mmol/L) LDL cholesterol (mmol/L) Triglycerides (mmol/L)
54.9 56 80.7 162.4 101.4 72.6
52.3 55 79.1 155.0 101.3 70.5
64 24 5.15 6.23 4.01 1.66
69 11 5.16 6.02 3.80 1.48
Mean where not otherwise stated.
ine (n 5 108) or felodipine (n 5 108). The characteristics of the patient groups are given in Table 1. Design During a washout phase all patients were treated with placebo for 4 weeks. Patients fulfilling the above BP criteria were entered into an 8-week dose titration phase and randomized either to amlodipine or felodipine ER. Initial dose during the titration phase was 5 mg for both amlodipine and felodipine. Patients who had not reached the target BP defined as a mean supine diastolic BP #90 mm Hg (5 response criterium) after 4 weeks had their dose increased to 10 mg of the respective drug. All patients were followed for another 4 weeks whether or not the dose had been increased. Tablet dummies were not used in the trial and thus patients were not blinded with respect to the design of the tablets. However, neither the patients nor the investigators were informed of the name of the medication, which was delivered to the patient in a closed box identical for both treatments. Therefore, in practically all cases the investigator was blinded regarding the study medication (the only exception was if a patient showed the tablets to the investigator). The same can be considered true for the patients, as very few of them had previously been treated with any of the study drugs. The patients were instructed to take their tablets once daily at the same time each morning. Blood Pressure Ambulatory automatic blood pressure measurements were scheduled for every 20 min during day and night using a SpaceLabs (Redmond, WA) model 90202 or 90207. The ABPM was performed after 4 weeks of placebo medication at the pretrial visit (week 0), after 4 weeks (week 4) on 5 mg of either drug, and at the week 8 (week 8) visit. Clinic BP was measured 23 to 25 h after previous drug intake. The BP was recorded twice with a 1-min
692
¨ STERGREN ET AL O
AJH–JUNE 1998 –VOL. 11, NO. 6, PART 1
TABLE 2. CHANGES OF THE 24-H AMBULATORY BLOOD PRESSURE BETWEEN WEEK 0 AND WEEK 4
Source of Reading* All 24 h SBP DBP Day SBP DBP Night SBP DBP Last 4 h SBP DBP
Number of Evaluable Patients A
F
Adjusted Difference A to F (mm Hg)†
Approximate 95% CI (mm Hg)
P
99 99
100 100
22.6 21.7
62.6 61.7
.060 .052
101 101
102 102
21.4 20.9
62.7 61.8
.29 .33
85 85
88 88
24.2 22.7
63.6 62.2
.026 .019
70 70
77 77
22.1 22.8
64.2 62.9
.33 .064
* “All 24 h”, 07:00 to 06:00, at least 19 readings; “Day”, 07:00 to 23:00, at least 9 readings; “Night”, 23:00 to 06:00, at least 8 readings; “Last 4 h”, start 120:00 to start 124:00. † Data adjusted for baseline values at visit week 0. The number of patients during “All 24 h” is lower than during “Day” as some of the patients turned off their equipment before 19 readings were achieved (which was required to be included in the all 24-h period). Most of the patients who stopped their equipment did so during the night, which resulted in a lower number of available patients during “Night” and during the “Last 4 h” readings. A, amlodipine; DBP, diastolic blood pressure; F, felodipine; SBP, systolic blood pressure.
interval after 5 min rest in the supine position and the mean value was used for analyses. The study nurse measured the BP throughout the study and used the same equipment and arm every time. Heart rate was measured both in the supine and the standing position during 30 sec after BP measurements. Plasma Concentrations Venous blood samples for plasma concentrations of the drugs were taken 23 to 25 h after dosing on the same day as the 24-h ABPMs were performed. The analyses were made by the Department of Clinical Pharmacology, Faculty of Health Sciences, Linko¨ping. Adverse Events All adverse events were recorded on the Case Report Form. The date of onset, duration, severity (mild, moderate, or severe), the most likely cause, and the outcome were stated. Adverse events were assessed at week 22, 0, 4, and 8 by using a general nonspecific question “How do you feel?” Ankle circumference, in millimeters, was measured in the morning 5 cm above the lower limit of the medial malleolus. Plasma Lipids Patients were instructed to fast overnight before the pretrial visit, the week 0, and the week 8 visits. Venous blood samples were taken in the morning. Total, LDL and HDL cholesterol and triglycerides were analyzed at a central laboratory (CALAB, Stockholm, Sweden). Statistical Analysis Blood pressure-lowering effect after 4 weeks was analyzed with respect to response/
nonresponse using x2 test and with respect to BPlowering effect using analysis of covariance with or without baseline pressure as covariate. After 8 weeks, the treatment groups were compared by x2 test with respect to response to low/high dose. For changes in ankle circumference, the KruskalWallis test was used. For comparison of the effects with respect to the lipids, an analysis of variance/ covariance was used with the factors treatment and dose. The baseline value was included as a covariate in the model when appropriate. Adverse experiences were evaluated with the x2 test or Fisher’s exact test when appropriate. Comparison of 24-h ABPM The mean levels of all 24-h, day, and night ABPM were calculated when at least 19, 9, and 8 valid readings were registered. The morning mean ABPM was calculated if at least five readings could be achieved during the last 4 h. As some patients stopped their equipment before the end of the 24-h period, the “last 4-h value” might be calculated from just 2 h. To illustrate the development over 24 h and achieve comparative values hour by hour, a moving average technique was used. A moving average based on five readings (MAV5) was calculated as: MAV5(i12) 5 (xi 1 xi11 1 xi12 1 . . . xi14)/5, i 5 1, 2 . . . n 2 4.
AJH–JUNE 1998 –VOL. 11, NO. 6, PART 1
24-HOUR BLOOD PRESSURE WITH AMLODIPINE V FELODIPINE
693
FIGURE 1. Hourly BP averages, moving averages based on five readings, constructed from the results of the 24-h ABPM for patients treated with amlodipine and felodipine ER. End points are based on three weighted readings at the start of 24-h ABPM and at the end of 24-h ABPM.
After calculating MAV5 the points in time remain unchanged. Therefore, they were unevenly distributed over time and do not agree when we wanted to compare the sequences of readings from two occasions. To get a reasonable representative level over a suitable time period (eg, 1 h), we had to adjust these MAV5 so that they can be compared hour by hour. A pragmatic way of doing this is to calculate a mean of MAV5 found within a particular hour, eg, 11:30 to 12:30, which will then be a level representing 12 h. In such a way we obtain levels for 8 h, 9 h, . . . 24 h, until 08:00 in the morning after (or any other 24 h sequence). Values at both ends were calculated from three readings.
amlodipine group. Past history of antihypertensive drug therapy was equally common in both groups. Approximately 50% in both groups had been treated within the past 6 months. Both treatment groups showed similar patterns of number of drugs used and median duration of treatment. The most commonly used drugs before entering the study were atenolol and metoprolol. Only four patients had previously been treated with dihydropyridine calcium antagonists.
TABLE 3. CHANGES OF CLINIC BPS AND HEART RATES BETWEEN WEEK 0 AND WEEK 4
Ethical Approval The study was approved by the ethical committee at the Karolinska Institute. RESULTS The treatment groups were similar with respect to the majority of the baseline factors. Table 1 shows a similar distribution of gender, age, weight, diastolic BP, heart rate, and lipid levels. A difference was seen in systolic BP, but was accounted for in the statistical analysis. Current smokers were more common in the
P Amlodipine v Felodipine ER Amlodipine Felodipine Changes (in mm Hg and beats/min)*
Source Diastolic BP, supine Systolic BP, supine Heart rate, supine
28.8 211.1 0.62
27.0 28.8 1.17
Mean values measured 24 h after drug intake. * Data adjusted for baseline values.
.024 .165 .595
694
¨ STERGREN ET AL O
AJH–JUNE 1998 –VOL. 11, NO. 6, PART 1
FIGURE 2. Box and whisker plot (median, 25% to 75%, 10% to 90%, and outliers) for clinic diastolic BPs at weeks 0, 4, and 8. A, amlodipine; F, felodipine.
Blood Pressure After 4 Weeks of Active Treatment Table 2 and Figure 1 show the changes in the 24-h ABPM after 4 weeks of treatment. Both amlodipine and felodipine induced a decrease in systolic as well as diastolic BP after 4 weeks of treatment with 5 mg once daily. Nighttime systolic and diastolic BP (23:00 to 06:00) was significantly (P 5 .026 and .019, respectively) more reduced by amlodipine than by felodipine. During the last 4 h of the 24-h ABPM period, amlodipine lowered diastolic BP 2.8 mm Hg more as compared to felodipine (P 5 .064). Table 3 and Figure 2 present the changes in clinic blood pressure and heart rate after 4 weeks of treatment. Amlodipine induced a significantly (P 5 .02) greater decrease in diastolic BP than did felodipine. No significant difference in heart rate was found between treatments. Fifty-two patients (50%) in the amlodipine group reached their target BP after 4 weeks as compared to 35 (33%) in the felodipine group (P 5 .013). Fifty-one patients (49%) in the amlodipine group and 71 (68%) patients in the felodipine group had their dose increased after 4 weeks of treatment.
FIGURE 3. Box and whisker plot (median, 25% to 75%, 10% to 90%, and outliers) for plasma concentrations of drugs at week 8. A, amlodipine; F, felodipine.
Blood Pressure After 8 Weeks of Active Treatment After 8 weeks of treatment, 84 patients (82%) treated with amlodipine had reached their target BP as compared to 71 patients (69%) treated with felodipine (P 5 .036; Table 4). Amlodipine reduced clinic BP slightly more than did felodipine. The difference in supine systolic BP was statistically significant (3.7 mm Hg; P 5 .031). The ABPM results showed a tendency to a better lowering effect on diastolic BP during night in the amlodipinetreated group (P 5 .08). No significant changes of plasma lipids, ankle circumference, or weight were found with either treatment. Plasma Concentrations The plasma concentration of the drug 24 h after dosing was considerably higher in the amlodipine group than in the felodipine group in which the concentration was zero in many cases. The plasma concentration level seemed to increase proportionally with dose in the amlodipine group but not in the felodipine group (Figure 3 and Table 5).
TABLE 4. TARGET BP AND RESPONDERS AT HIGH/LOW DOSE AT WEEK 8 No. of Evaluable Patients Source
Yes
Yes
Amlodipine
Felodipine
Amlodipine
Felodipine
P
Target BP
103
103
103
103
Responder, high dose
103
103
Nonresponder
103
103
71 (69%) 30 (27%) 41 (40%) 32 (31%)
.036
Responder, low dose
84 (82%) 45 (44%) 39 (38%) 19 (18%)
.042
AJH–JUNE 1998 –VOL. 11, NO. 6, PART 1
695
24-HOUR BLOOD PRESSURE WITH AMLODIPINE V FELODIPINE
TABLE 5. PLASMA CONCENTRATION OF DRUGS AFTER 4 AND 8 WEEKS OF ACTIVE TREATMENT FOR ALL PATIENTS AND FOR PATIENTS DIVIDED INTO LOW (5 MG) AND HIGH DOSE (10 MG) Number of Evaluable Patients Amlodipine
Plasma Concentration (ng/mL)
Felodipine
Amlodipine
Felodipine
Source
Low
High
Low
High
Low
High
Low
High
Week 4 Week 8
51 49
51 48
28 26
64 61
5.7 6.1
6.0 13.1
0.3 0.4
0.4 0.7
Values are median.
Adverse Events The most common side effects were edema and headache, with similar occurrences in both groups. A statistically significant difference between the treatments in favor of amlodipine could be demonstrated only with respect to the number of patients with severe adverse events (P 5 .018; Table 6). DISCUSSION When comparing the effect after 4 weeks of active treatment, amlodipine appeared to be more effective than felodipine in controlling nighttime and morning BP when the drugs were taken as single doses of 5 mg once daily in the morning. Nighttime systolic and diastolic BP was significantly more reduced by amlodipine than by felodipine. During the last 4 h of the ABPM amlodipine reduced diastolic BP 2.8 mm Hg more than felodipine. The validity of the ABPM findings is strengthened by the fact that the amlodipine treatment showed a significantly greater decrease in BP also when conventional techniques were used to assess the effect 24 h after dosing. A significantly greater number of patients in the amlodipine group thus reached their target BP and fewer patients needed to have their dose increased at the 4-week assessment, as compared to the patients in
TABLE 6. ADVERSE EVENT WITH AN ONSET DURING ACTIVE TREATMENT Amlodipine Felodipine (n 5 107) (n 5 107) Patients with adverse events Adverse events Patients with severe adverse events Patients who discontinued due to adverse events Patients with dose reduced or temporarily discontinued due to adverse events
P
45 79
52 93
.34 NS
2
10
.018
5
10
.18
4
2
.68
“Severe adverse events” indicate events that are considered to be severe by the clinical trialist (scale: mild, moderate, severe).
the felodipine group. The design of the study in this way precluded a direct comparative evaluation after 8 weeks’ treatment as the decision to give a low or high dose was based on the clinic BP at 4 weeks. Therefore, the analyses after 8 weeks must be regarded as explorative rather than confirmative. Despite the fact that more patients on felodipine had a higher dose at 8 weeks, there were still more responders in the group treated with amlodipine (82% v 69%) and nighttime ABPM also tended to be lower on treatment with amlodipine. Thus, in all measurements, nighttime and morning BP were slightly lower on amlodipine, whereas daytime BP was equally lowered by the two drugs. This is in accordance with the difference in pharmacokinetics of the drugs9 and also with the results of our analyses of plasma concentrations at the end of the dose interval, which showed absent or low values of felodipine. A good correlation between the clinical effect and plasma concentration of dihydropyridines has been reported earlier.7,10 –12 For felodipine ER a variation in peak-to-trough concentration exists and therefore, the extent of the antihypertensive effect varies over 24 h.8,13 Bainbridge et al14 in a comparative study found that trough-to-peak variability for plasma drug concentration was 37% for felodipine ER and 67% for amlodipine, corresponding to a lesser variability with amlodipine in the BP reduction across the dosage interval. It can be argued that the small difference in BPlowering effect between the treatments (2 to 3 mm Hg) during night and morning hours, although statistically significant, will not affect cardiovascular morbidity. This is certainly true for many individuals, but for the total hypertensive population even such a small difference in BP could induce a difference in outcome regarding cardiovascular morbidity. The effects of hypertension treatment seen in the major trials have been found with ,6 mm Hg lowering of the mean diastolic pressure in the actively treated groups.15 Thus, for individual patients for whom it is believed to be especially important to control nighttime and morning BP, amlodipine seems to be a more reliable alternative than felodipine ER.
696
¨ STERGREN ET AL O
The number of patients with reported adverse events tended to be higher in the felodipine group. This indicates that the slightly more effective hypotensive action of amlodipine was not achieved at the cost of increased side effects. Also this finding may be attributable to the pharmacodynamic profile of amlodipine with a slow onset of action, a lower plasma peak, and a longer duration of action.16,17 A superior relation between response rate and side effects of amlodipine in comparison with felodipine was reported by van der Krogt et al.18 Also Hoegholm et al19 reported a lower incidence of side effects with amlodipine as compared to felodipine ER. In that study the dose was up-titrated depending on response and the mean daily dose of felodipine was 11.2 mg and for amlodipine 7.4 mg, resulting in a similar diastolic BP decrease. Systolic 24-h ambulatory BP was significantly greater with amlodipine. These studies indicate that the pharmacodynamic profile may be of importance for the frequency of side effects. In our study, however, the most common side effects (headache and edema) did not differ between treatments, and measurements of ankle circumference yielded the same results in both groups (ie, unchanged circumference during both treatments). The time of measurement being in the morning, however, is not optimal to detect this side effect, as the occurrence of edema is more frequent during the later part of the day due to the higher ortohydrostatic pressure in the legs in the upright position during the day. In conclusion, both amlodipine and felodipine lower BP in patients with mild-to-moderate primary hypertension. When comparing the effect of the two drugs after 4 weeks of active treatment, ambulatory BP is equally lowered by both drugs during daytime, but is more effectively lowered by amlodipine during night and morning hours. Furthermore, amlodipine treatment induces a greater decrease of BP also when clinic BP is used to assess the effect 24 h after dose intake. The findings are compatible with the longer duration of action of amlodipine than of felodipine ER and may have therapeutic implications for the prevention of BP increases in the morning hours in patients at high risk of cardiovascular events.
AJH–JUNE 1998 –VOL. 11, NO. 6, PART 1
in the frequency of onset of acute myocardial infarction. N Engl J Med 1985;313:1315–1322. 4.
Muller JE, Ludmer PL, Willich SN, et al: Circadian variation in the frequency of onset of sudden cardiac death. Circulation 1987;75:131–138.
5.
Leenen, FHH: Clinical relevance of 24 h blood pressure control by 1,4-dihydropyridines. Am J Hypertens 1996; 9(suppl):97S–104S.
6.
Meredith PA, Elliot HL: Clinical pharmacokinetics of amlodipine. Clin Pharmacokinetics 1992;22:22–31.
7.
Edgar B, Collste P, Haglund K, et al: Pharmacokinetics and haemodynamic effects of felodipine as monotherapy in hypertensive patients. Clin Invest Med 1987; 10:388 –394.
8.
Fariello R, Boni E, Corda L, et al: Extended release felodipine in essential hypertension. Am J Hypertens 1991;4:17–33.
9.
Kelley JG, O’Malley K: Clinical pharmacokinetics of calcium antagonists: an update. Clin Pharmacokinet 1992;22:416 – 443.
10.
Kleinbloessem CH, van Brummelen P, Danhof M, et al: Rate-of-increase of plasma concentrations of nifedipine as a major determinant of its haemodynamic effect in humans. Clin Pharmacol Ther 1987;41:26 –30.
11.
Blychert E, Hedner T, Dahlo¨f C, Elmfeldt D: Plasma concentration effect relationships of intravenous and extended release oral felodipine in hypertensive patients. J Cardiovasc Pharmacol 1990;15:428 – 435.
12.
Donelly R, Meredith PA, Miller SHK, et al: Pharmacodynamic modeling of the antihypertensive response to amlodipine. Clin Pharmacol Ther 1993;54:303–310.
13.
Myers MG, Leenen FHH, Tanner J: Differential effects of felodipine and nifedipine on 24 hour blood pressure and left ventricular mass. Am J Hypertens 1995;8:712– 718.
14.
Bainbridge AD, Herlihy O, Meredith PA, Elliot HL: A comparative assessment of amlodipine and felodipine ER: pharmacokinetic and pharmacodynamic indices. Eur J Clin Pharmacol 1993;45:425– 430.
15.
MacMahon S, Peto R, Cutler J, et al: Blood pressure, stroke and coronary heart disease. Lancet 1990;335:765– 774.
16.
Heber ME, Brigden G, Al-Khawaja, Raftery EB: 24 h blood pressure control with the once daily calcium antagonist, amlodipine. Br J Clin Pharmacol 1989;27: 359 –365.
17.
Julius S: Amlodipine in hypertension: an overview of the clinical dossier. J Cardiovasc Pharmacol 1988;12: S27–S33.
18.
van der Krogt J, Brand R, Dawson EC: Amlodipine versus extended release felodipine in general practice: a randomized, parallel-group study in patients with mild-to-moderate hypertension. Curr Therapeutic Res 1996;57:145–158.
19.
Hoegholm A, Wiinberg N, Rasmussen E, et al: Comparative effects of amlodipine and felodipine ER on office and ambulatory blood pressure in patients with mild to moderate hypertension. J Human Hypertens 1995;9(suppl 1):S25–S28.
ACKNOWLEDGMENTS We are grateful to Ylva Falkeborn for valuable assistance, to Elisabeth Berg for statistical assistance, and to Johan Ahlner, Linko¨ping, for analyzing the plasma drug concentrations.
REFERENCES 1.
Millar-Craig MW, Bishop CN, Raftery EB: Circadian variation of blood pressure. Lancet 1978;i:795–797.
2.
Pickering TG: Ambulatory Monitoring and Blood Pressure Variability. Science Press, London, 1991.
3.
Muller JE, Stone PH, Turi ZG, et al: Circadian variation
1998;11:690 – 696
Effect of Amlodipine Versus Felodipine Extended Release on 24-Hour Ambulatory Blood Pressure in Hypertension ¨ stergren, Hans Isaksson, Ulf Brodin, Åke Schwan, and K. Peter O ¨ hman for the study group Jan O
Amlodipine and felodipine are calcium antagonists of the dihydropyridine type. The elimination halflife of amlodipine is longer than that of felodipine. To study whether the different elimination rates of the drugs were reflected in different duration of blood pressure (BP) control, we compared amlodipine and felodipine extended release (ER) by both conventional clinic BP 24 h after drug intake and 24 h ambulatory BP monitoring (ABPM), with special reference to nighttime and morning blood pressure. Two hundred and sixteen patients with primary hypertension (supine diastolic BP, 95 to 115 mm Hg) were randomized to receive amlodipine or felodipine ER in a multicenter study. The starting dose of both drugs was 5 mg. If the target clinic diastolic BP (90 mm Hg) had not been achieved after 4 weeks the dose was increased to 10 mg. Twenty-four-hour ABPM was performed with the subjects taking placebo medication before randomization and after 4 and 8 weeks undergoing active treatment.
B
Significantly more patients responded after 4 weeks of treatment with amlodipine (50%) as compared with felodipine (33%) (P 5 .013). ABPM during daytime (07:00 to 23:00) was similar during both treatments, but nighttime systolic (P 5 .026) and diastolic (P 5 .019) BP was more effectively reduced by amlodipine than by felodipine. After 8 weeks 82% achieved the target pressure with amlodipine and 69% with felodipine (P 5 .036 for the difference). Amlodipine seems to be more effective than felodipine when the drugs are compared in the same dose, with regard to the effect on clinic BP 24 h after dosing and to ambulatory BP during the night. The longer elimination half-life of amlodipine as compared to felodipine is the probable reason for this finding. Am J Hypertens 1998;11:690 – 696 © 1998 American Journal of Hypertension, Ltd. KEY WORDS:
Ambulatory blood pressure, circadian variation, amlodipine, felodipine, plasma half-life.
lood pressure (BP) increases rather rapidly during the early morning hours1,2 and in some cases the increase in BP may contribute to the increase in cardiovascular events seen during this time period.3,4 Therefore, it may be impor-
tant to have an adequate BP-lowering effect during the entire 24-h period to avoid pressure peaks at the end of the dose interval when treating hypertensive subjects with a drug that is given once daily.5 The dihydropyridine calcium antagonists amlodip-
Received April 23, 1997. Accepted January 20, 1998. ¨ ); DeFrom the Department of Medicine, Karolinska Hospital (JO partment of Medicine, So¨dersjukhuset (HI); Medical Statistics Unit, Karolinska Institute (UB), Stockholm; Department of Family Medicine, Uppsala University (AS); and the Department of Endocrinol¨ ), Linko¨ping, Sweden. ogy, University Hospital (KPO
This study was supported by Pfizer AB, Sweden. ¨ sterAddress for correspondence and reprint requests to Jan O gren, MD, PhD, Department of Medicine, Karolinska Hospital, 171 76 Stockholm, Sweden; e-mail: [email protected]
© 1998 by the American Journal of Hypertension, Ltd. Published by Elsevier Science, Inc.
0895-7061/98/$19.00 PII S0895-7061(98)00032-6
AJH–JUNE 1998 –VOL. 11, NO. 6, PART 1
24-HOUR BLOOD PRESSURE WITH AMLODIPINE V FELODIPINE
ine and felodipine are frequently used in the treatment of hypertension. Peak plasma level for amlodipine occurs 6 to 12 h after oral tablet intake, with an elimination half-life of 35 to 50 h.6 For felodipine, the peak plasma concentration occurs after 1 to 3 h and the elimination half-life is 6 to 8 h.7 With felodipine as an extended release (ER) formulation, the time to peak plasma concentration is prolonged to 3 to 7 h after dosing and the plasma concentration at 24 h after dosing is increased.7,8 The primary objective of this study was to investigate whether starting doses of 5 mg of amlodipine versus 5 mg of felodipine ER are equipotent in the treatment of primary hypertension with focus on the antihypertensive effect at the end of the dosing interval. Twenty-four-hour ambulatory BP monitoring (ABPM) was used to allow analysis of the BP-lowering effect of the drugs in the later part of the dosing interval (ie, during the night and the early morning hours). The primary efficacy variable was the decrease in mean diastolic BP in the last 4 h in the 24-h ABPMs after 4 weeks of treatment with 5 mg of either drug. Secondary objectives were 1) to investigate the effect on 24-h ABPM of uptitrated doses to 10 mg when necessary, 2) to investigate the effect on clinic BP, 3) to investigate subjective side effects and ankle edema, and 4) to investigate the effect of treatment on fasting plasma lipids. In addition, plasma concentrations of the drugs were measured in conjunction with clinic BP at the end of the dose interval. PATIENTS AND METHODS The study was a single-blind, multicenter, randomized, parallel group study. Patients with mild-to-moderate primary hypertension (mean supine diastolic BP, 95 to 115 mm Hg) were enrolled at 13 centers in Sweden by 18 investigators. Patients were excluded if they had insulin-treated diabetes mellitus, renal impairment, or if they had experienced a cerebrovascular accident within the past 3 months and angina or coronary infarction within the past 6 months. Medication that might alter BP and changes in lipid-lowering treatment were not allowed in the study. Patients who had previously been treated with antihypertensive drugs had their medication withdrawn before entering the study. The study was initially planned for 140 patients to complete the study. One hundred and sixty patients were initially enrolled. A preplanned interim statistical analysis of the variation in BP changes during the last 4 h of the recording was performed to investigate the power of the study. It was then decided that an additional 56 patients were to be enrolled. Thus, 216 patients (120 men and 96 women) were entered into the study and randomized to treatment with amlodip-
691
TABLE 1. BASELINE DATA FOR ALL PATIENTS, FROM VISIT WEEK 0 Factor
Amlodipine
Felodipine
Age (years) Gender (percent men) Weight (kg) Systolic BP, supine (mm Hg) Diastolic BP, supine (mm Hg) Heart rate, supine Previous treatment for hypertension (%) Smokers (no.) Fasting glucose (mmol/L) Total cholesterol (mmol/L) LDL cholesterol (mmol/L) Triglycerides (mmol/L)
54.9 56 80.7 162.4 101.4 72.6
52.3 55 79.1 155.0 101.3 70.5
64 24 5.15 6.23 4.01 1.66
69 11 5.16 6.02 3.80 1.48
Mean where not otherwise stated.
ine (n 5 108) or felodipine (n 5 108). The characteristics of the patient groups are given in Table 1. Design During a washout phase all patients were treated with placebo for 4 weeks. Patients fulfilling the above BP criteria were entered into an 8-week dose titration phase and randomized either to amlodipine or felodipine ER. Initial dose during the titration phase was 5 mg for both amlodipine and felodipine. Patients who had not reached the target BP defined as a mean supine diastolic BP #90 mm Hg (5 response criterium) after 4 weeks had their dose increased to 10 mg of the respective drug. All patients were followed for another 4 weeks whether or not the dose had been increased. Tablet dummies were not used in the trial and thus patients were not blinded with respect to the design of the tablets. However, neither the patients nor the investigators were informed of the name of the medication, which was delivered to the patient in a closed box identical for both treatments. Therefore, in practically all cases the investigator was blinded regarding the study medication (the only exception was if a patient showed the tablets to the investigator). The same can be considered true for the patients, as very few of them had previously been treated with any of the study drugs. The patients were instructed to take their tablets once daily at the same time each morning. Blood Pressure Ambulatory automatic blood pressure measurements were scheduled for every 20 min during day and night using a SpaceLabs (Redmond, WA) model 90202 or 90207. The ABPM was performed after 4 weeks of placebo medication at the pretrial visit (week 0), after 4 weeks (week 4) on 5 mg of either drug, and at the week 8 (week 8) visit. Clinic BP was measured 23 to 25 h after previous drug intake. The BP was recorded twice with a 1-min
692
¨ STERGREN ET AL O
AJH–JUNE 1998 –VOL. 11, NO. 6, PART 1
TABLE 2. CHANGES OF THE 24-H AMBULATORY BLOOD PRESSURE BETWEEN WEEK 0 AND WEEK 4
Source of Reading* All 24 h SBP DBP Day SBP DBP Night SBP DBP Last 4 h SBP DBP
Number of Evaluable Patients A
F
Adjusted Difference A to F (mm Hg)†
Approximate 95% CI (mm Hg)
P
99 99
100 100
22.6 21.7
62.6 61.7
.060 .052
101 101
102 102
21.4 20.9
62.7 61.8
.29 .33
85 85
88 88
24.2 22.7
63.6 62.2
.026 .019
70 70
77 77
22.1 22.8
64.2 62.9
.33 .064
* “All 24 h”, 07:00 to 06:00, at least 19 readings; “Day”, 07:00 to 23:00, at least 9 readings; “Night”, 23:00 to 06:00, at least 8 readings; “Last 4 h”, start 120:00 to start 124:00. † Data adjusted for baseline values at visit week 0. The number of patients during “All 24 h” is lower than during “Day” as some of the patients turned off their equipment before 19 readings were achieved (which was required to be included in the all 24-h period). Most of the patients who stopped their equipment did so during the night, which resulted in a lower number of available patients during “Night” and during the “Last 4 h” readings. A, amlodipine; DBP, diastolic blood pressure; F, felodipine; SBP, systolic blood pressure.
interval after 5 min rest in the supine position and the mean value was used for analyses. The study nurse measured the BP throughout the study and used the same equipment and arm every time. Heart rate was measured both in the supine and the standing position during 30 sec after BP measurements. Plasma Concentrations Venous blood samples for plasma concentrations of the drugs were taken 23 to 25 h after dosing on the same day as the 24-h ABPMs were performed. The analyses were made by the Department of Clinical Pharmacology, Faculty of Health Sciences, Linko¨ping. Adverse Events All adverse events were recorded on the Case Report Form. The date of onset, duration, severity (mild, moderate, or severe), the most likely cause, and the outcome were stated. Adverse events were assessed at week 22, 0, 4, and 8 by using a general nonspecific question “How do you feel?” Ankle circumference, in millimeters, was measured in the morning 5 cm above the lower limit of the medial malleolus. Plasma Lipids Patients were instructed to fast overnight before the pretrial visit, the week 0, and the week 8 visits. Venous blood samples were taken in the morning. Total, LDL and HDL cholesterol and triglycerides were analyzed at a central laboratory (CALAB, Stockholm, Sweden). Statistical Analysis Blood pressure-lowering effect after 4 weeks was analyzed with respect to response/
nonresponse using x2 test and with respect to BPlowering effect using analysis of covariance with or without baseline pressure as covariate. After 8 weeks, the treatment groups were compared by x2 test with respect to response to low/high dose. For changes in ankle circumference, the KruskalWallis test was used. For comparison of the effects with respect to the lipids, an analysis of variance/ covariance was used with the factors treatment and dose. The baseline value was included as a covariate in the model when appropriate. Adverse experiences were evaluated with the x2 test or Fisher’s exact test when appropriate. Comparison of 24-h ABPM The mean levels of all 24-h, day, and night ABPM were calculated when at least 19, 9, and 8 valid readings were registered. The morning mean ABPM was calculated if at least five readings could be achieved during the last 4 h. As some patients stopped their equipment before the end of the 24-h period, the “last 4-h value” might be calculated from just 2 h. To illustrate the development over 24 h and achieve comparative values hour by hour, a moving average technique was used. A moving average based on five readings (MAV5) was calculated as: MAV5(i12) 5 (xi 1 xi11 1 xi12 1 . . . xi14)/5, i 5 1, 2 . . . n 2 4.
AJH–JUNE 1998 –VOL. 11, NO. 6, PART 1
24-HOUR BLOOD PRESSURE WITH AMLODIPINE V FELODIPINE
693
FIGURE 1. Hourly BP averages, moving averages based on five readings, constructed from the results of the 24-h ABPM for patients treated with amlodipine and felodipine ER. End points are based on three weighted readings at the start of 24-h ABPM and at the end of 24-h ABPM.
After calculating MAV5 the points in time remain unchanged. Therefore, they were unevenly distributed over time and do not agree when we wanted to compare the sequences of readings from two occasions. To get a reasonable representative level over a suitable time period (eg, 1 h), we had to adjust these MAV5 so that they can be compared hour by hour. A pragmatic way of doing this is to calculate a mean of MAV5 found within a particular hour, eg, 11:30 to 12:30, which will then be a level representing 12 h. In such a way we obtain levels for 8 h, 9 h, . . . 24 h, until 08:00 in the morning after (or any other 24 h sequence). Values at both ends were calculated from three readings.
amlodipine group. Past history of antihypertensive drug therapy was equally common in both groups. Approximately 50% in both groups had been treated within the past 6 months. Both treatment groups showed similar patterns of number of drugs used and median duration of treatment. The most commonly used drugs before entering the study were atenolol and metoprolol. Only four patients had previously been treated with dihydropyridine calcium antagonists.
TABLE 3. CHANGES OF CLINIC BPS AND HEART RATES BETWEEN WEEK 0 AND WEEK 4
Ethical Approval The study was approved by the ethical committee at the Karolinska Institute. RESULTS The treatment groups were similar with respect to the majority of the baseline factors. Table 1 shows a similar distribution of gender, age, weight, diastolic BP, heart rate, and lipid levels. A difference was seen in systolic BP, but was accounted for in the statistical analysis. Current smokers were more common in the
P Amlodipine v Felodipine ER Amlodipine Felodipine Changes (in mm Hg and beats/min)*
Source Diastolic BP, supine Systolic BP, supine Heart rate, supine
28.8 211.1 0.62
27.0 28.8 1.17
Mean values measured 24 h after drug intake. * Data adjusted for baseline values.
.024 .165 .595
694
¨ STERGREN ET AL O
AJH–JUNE 1998 –VOL. 11, NO. 6, PART 1
FIGURE 2. Box and whisker plot (median, 25% to 75%, 10% to 90%, and outliers) for clinic diastolic BPs at weeks 0, 4, and 8. A, amlodipine; F, felodipine.
Blood Pressure After 4 Weeks of Active Treatment Table 2 and Figure 1 show the changes in the 24-h ABPM after 4 weeks of treatment. Both amlodipine and felodipine induced a decrease in systolic as well as diastolic BP after 4 weeks of treatment with 5 mg once daily. Nighttime systolic and diastolic BP (23:00 to 06:00) was significantly (P 5 .026 and .019, respectively) more reduced by amlodipine than by felodipine. During the last 4 h of the 24-h ABPM period, amlodipine lowered diastolic BP 2.8 mm Hg more as compared to felodipine (P 5 .064). Table 3 and Figure 2 present the changes in clinic blood pressure and heart rate after 4 weeks of treatment. Amlodipine induced a significantly (P 5 .02) greater decrease in diastolic BP than did felodipine. No significant difference in heart rate was found between treatments. Fifty-two patients (50%) in the amlodipine group reached their target BP after 4 weeks as compared to 35 (33%) in the felodipine group (P 5 .013). Fifty-one patients (49%) in the amlodipine group and 71 (68%) patients in the felodipine group had their dose increased after 4 weeks of treatment.
FIGURE 3. Box and whisker plot (median, 25% to 75%, 10% to 90%, and outliers) for plasma concentrations of drugs at week 8. A, amlodipine; F, felodipine.
Blood Pressure After 8 Weeks of Active Treatment After 8 weeks of treatment, 84 patients (82%) treated with amlodipine had reached their target BP as compared to 71 patients (69%) treated with felodipine (P 5 .036; Table 4). Amlodipine reduced clinic BP slightly more than did felodipine. The difference in supine systolic BP was statistically significant (3.7 mm Hg; P 5 .031). The ABPM results showed a tendency to a better lowering effect on diastolic BP during night in the amlodipinetreated group (P 5 .08). No significant changes of plasma lipids, ankle circumference, or weight were found with either treatment. Plasma Concentrations The plasma concentration of the drug 24 h after dosing was considerably higher in the amlodipine group than in the felodipine group in which the concentration was zero in many cases. The plasma concentration level seemed to increase proportionally with dose in the amlodipine group but not in the felodipine group (Figure 3 and Table 5).
TABLE 4. TARGET BP AND RESPONDERS AT HIGH/LOW DOSE AT WEEK 8 No. of Evaluable Patients Source
Yes
Yes
Amlodipine
Felodipine
Amlodipine
Felodipine
P
Target BP
103
103
103
103
Responder, high dose
103
103
Nonresponder
103
103
71 (69%) 30 (27%) 41 (40%) 32 (31%)
.036
Responder, low dose
84 (82%) 45 (44%) 39 (38%) 19 (18%)
.042
AJH–JUNE 1998 –VOL. 11, NO. 6, PART 1
695
24-HOUR BLOOD PRESSURE WITH AMLODIPINE V FELODIPINE
TABLE 5. PLASMA CONCENTRATION OF DRUGS AFTER 4 AND 8 WEEKS OF ACTIVE TREATMENT FOR ALL PATIENTS AND FOR PATIENTS DIVIDED INTO LOW (5 MG) AND HIGH DOSE (10 MG) Number of Evaluable Patients Amlodipine
Plasma Concentration (ng/mL)
Felodipine
Amlodipine
Felodipine
Source
Low
High
Low
High
Low
High
Low
High
Week 4 Week 8
51 49
51 48
28 26
64 61
5.7 6.1
6.0 13.1
0.3 0.4
0.4 0.7
Values are median.
Adverse Events The most common side effects were edema and headache, with similar occurrences in both groups. A statistically significant difference between the treatments in favor of amlodipine could be demonstrated only with respect to the number of patients with severe adverse events (P 5 .018; Table 6). DISCUSSION When comparing the effect after 4 weeks of active treatment, amlodipine appeared to be more effective than felodipine in controlling nighttime and morning BP when the drugs were taken as single doses of 5 mg once daily in the morning. Nighttime systolic and diastolic BP was significantly more reduced by amlodipine than by felodipine. During the last 4 h of the ABPM amlodipine reduced diastolic BP 2.8 mm Hg more than felodipine. The validity of the ABPM findings is strengthened by the fact that the amlodipine treatment showed a significantly greater decrease in BP also when conventional techniques were used to assess the effect 24 h after dosing. A significantly greater number of patients in the amlodipine group thus reached their target BP and fewer patients needed to have their dose increased at the 4-week assessment, as compared to the patients in
TABLE 6. ADVERSE EVENT WITH AN ONSET DURING ACTIVE TREATMENT Amlodipine Felodipine (n 5 107) (n 5 107) Patients with adverse events Adverse events Patients with severe adverse events Patients who discontinued due to adverse events Patients with dose reduced or temporarily discontinued due to adverse events
P
45 79
52 93
.34 NS
2
10
.018
5
10
.18
4
2
.68
“Severe adverse events” indicate events that are considered to be severe by the clinical trialist (scale: mild, moderate, severe).
the felodipine group. The design of the study in this way precluded a direct comparative evaluation after 8 weeks’ treatment as the decision to give a low or high dose was based on the clinic BP at 4 weeks. Therefore, the analyses after 8 weeks must be regarded as explorative rather than confirmative. Despite the fact that more patients on felodipine had a higher dose at 8 weeks, there were still more responders in the group treated with amlodipine (82% v 69%) and nighttime ABPM also tended to be lower on treatment with amlodipine. Thus, in all measurements, nighttime and morning BP were slightly lower on amlodipine, whereas daytime BP was equally lowered by the two drugs. This is in accordance with the difference in pharmacokinetics of the drugs9 and also with the results of our analyses of plasma concentrations at the end of the dose interval, which showed absent or low values of felodipine. A good correlation between the clinical effect and plasma concentration of dihydropyridines has been reported earlier.7,10 –12 For felodipine ER a variation in peak-to-trough concentration exists and therefore, the extent of the antihypertensive effect varies over 24 h.8,13 Bainbridge et al14 in a comparative study found that trough-to-peak variability for plasma drug concentration was 37% for felodipine ER and 67% for amlodipine, corresponding to a lesser variability with amlodipine in the BP reduction across the dosage interval. It can be argued that the small difference in BPlowering effect between the treatments (2 to 3 mm Hg) during night and morning hours, although statistically significant, will not affect cardiovascular morbidity. This is certainly true for many individuals, but for the total hypertensive population even such a small difference in BP could induce a difference in outcome regarding cardiovascular morbidity. The effects of hypertension treatment seen in the major trials have been found with ,6 mm Hg lowering of the mean diastolic pressure in the actively treated groups.15 Thus, for individual patients for whom it is believed to be especially important to control nighttime and morning BP, amlodipine seems to be a more reliable alternative than felodipine ER.
696
¨ STERGREN ET AL O
The number of patients with reported adverse events tended to be higher in the felodipine group. This indicates that the slightly more effective hypotensive action of amlodipine was not achieved at the cost of increased side effects. Also this finding may be attributable to the pharmacodynamic profile of amlodipine with a slow onset of action, a lower plasma peak, and a longer duration of action.16,17 A superior relation between response rate and side effects of amlodipine in comparison with felodipine was reported by van der Krogt et al.18 Also Hoegholm et al19 reported a lower incidence of side effects with amlodipine as compared to felodipine ER. In that study the dose was up-titrated depending on response and the mean daily dose of felodipine was 11.2 mg and for amlodipine 7.4 mg, resulting in a similar diastolic BP decrease. Systolic 24-h ambulatory BP was significantly greater with amlodipine. These studies indicate that the pharmacodynamic profile may be of importance for the frequency of side effects. In our study, however, the most common side effects (headache and edema) did not differ between treatments, and measurements of ankle circumference yielded the same results in both groups (ie, unchanged circumference during both treatments). The time of measurement being in the morning, however, is not optimal to detect this side effect, as the occurrence of edema is more frequent during the later part of the day due to the higher ortohydrostatic pressure in the legs in the upright position during the day. In conclusion, both amlodipine and felodipine lower BP in patients with mild-to-moderate primary hypertension. When comparing the effect of the two drugs after 4 weeks of active treatment, ambulatory BP is equally lowered by both drugs during daytime, but is more effectively lowered by amlodipine during night and morning hours. Furthermore, amlodipine treatment induces a greater decrease of BP also when clinic BP is used to assess the effect 24 h after dose intake. The findings are compatible with the longer duration of action of amlodipine than of felodipine ER and may have therapeutic implications for the prevention of BP increases in the morning hours in patients at high risk of cardiovascular events.
AJH–JUNE 1998 –VOL. 11, NO. 6, PART 1
in the frequency of onset of acute myocardial infarction. N Engl J Med 1985;313:1315–1322. 4.
Muller JE, Ludmer PL, Willich SN, et al: Circadian variation in the frequency of onset of sudden cardiac death. Circulation 1987;75:131–138.
5.
Leenen, FHH: Clinical relevance of 24 h blood pressure control by 1,4-dihydropyridines. Am J Hypertens 1996; 9(suppl):97S–104S.
6.
Meredith PA, Elliot HL: Clinical pharmacokinetics of amlodipine. Clin Pharmacokinetics 1992;22:22–31.
7.
Edgar B, Collste P, Haglund K, et al: Pharmacokinetics and haemodynamic effects of felodipine as monotherapy in hypertensive patients. Clin Invest Med 1987; 10:388 –394.
8.
Fariello R, Boni E, Corda L, et al: Extended release felodipine in essential hypertension. Am J Hypertens 1991;4:17–33.
9.
Kelley JG, O’Malley K: Clinical pharmacokinetics of calcium antagonists: an update. Clin Pharmacokinet 1992;22:416 – 443.
10.
Kleinbloessem CH, van Brummelen P, Danhof M, et al: Rate-of-increase of plasma concentrations of nifedipine as a major determinant of its haemodynamic effect in humans. Clin Pharmacol Ther 1987;41:26 –30.
11.
Blychert E, Hedner T, Dahlo¨f C, Elmfeldt D: Plasma concentration effect relationships of intravenous and extended release oral felodipine in hypertensive patients. J Cardiovasc Pharmacol 1990;15:428 – 435.
12.
Donelly R, Meredith PA, Miller SHK, et al: Pharmacodynamic modeling of the antihypertensive response to amlodipine. Clin Pharmacol Ther 1993;54:303–310.
13.
Myers MG, Leenen FHH, Tanner J: Differential effects of felodipine and nifedipine on 24 hour blood pressure and left ventricular mass. Am J Hypertens 1995;8:712– 718.
14.
Bainbridge AD, Herlihy O, Meredith PA, Elliot HL: A comparative assessment of amlodipine and felodipine ER: pharmacokinetic and pharmacodynamic indices. Eur J Clin Pharmacol 1993;45:425– 430.
15.
MacMahon S, Peto R, Cutler J, et al: Blood pressure, stroke and coronary heart disease. Lancet 1990;335:765– 774.
16.
Heber ME, Brigden G, Al-Khawaja, Raftery EB: 24 h blood pressure control with the once daily calcium antagonist, amlodipine. Br J Clin Pharmacol 1989;27: 359 –365.
17.
Julius S: Amlodipine in hypertension: an overview of the clinical dossier. J Cardiovasc Pharmacol 1988;12: S27–S33.
18.
van der Krogt J, Brand R, Dawson EC: Amlodipine versus extended release felodipine in general practice: a randomized, parallel-group study in patients with mild-to-moderate hypertension. Curr Therapeutic Res 1996;57:145–158.
19.
Hoegholm A, Wiinberg N, Rasmussen E, et al: Comparative effects of amlodipine and felodipine ER on office and ambulatory blood pressure in patients with mild to moderate hypertension. J Human Hypertens 1995;9(suppl 1):S25–S28.
ACKNOWLEDGMENTS We are grateful to Ylva Falkeborn for valuable assistance, to Elisabeth Berg for statistical assistance, and to Johan Ahlner, Linko¨ping, for analyzing the plasma drug concentrations.
REFERENCES 1.
Millar-Craig MW, Bishop CN, Raftery EB: Circadian variation of blood pressure. Lancet 1978;i:795–797.
2.
Pickering TG: Ambulatory Monitoring and Blood Pressure Variability. Science Press, London, 1991.
3.
Muller JE, Stone PH, Turi ZG, et al: Circadian variation