Comparison of Nifedipine Alone and With Diltiazem or Verapamil in Hypertension
Receptor binding studies suggest that combinations of calcium channel blockers may result in either enhanced or diminished pharmacological effects, but clinical data in hypertension are incomplete. In this study, we compared blood pressure reductions using nifedipine alone, nifedipine plus diltiazem, and nifedipine plus verapamil and determined whether combinations alter nifedipine pharmacokinetics. After determination of baseline blood pressures, 16 subjects with essential hypertension (12 men, 4 women; mean age, 48 years) received 30 mg/d open-label, sustained release nifedipine for 2 weeks. If still hypertensive (n=16), they were randomized (double-blind) to receive either additional sustained release diltiazem or sustained release verapamil, both 180 mg/d, for 2 weeks and were then crossed-over for the final 2 weeks of the study. All medications were once-daily, extended-release formulations. Blood pressures and nifedipine plasma concentrations were measured during the final day of each treatment. Overall, each combination lowered mean systolic and diastolic pressures more than nifedipine alone. Mean supine diastolic pressures were significantly lower at 8 hours (77.6 versus 84.6 mm Hg, P=.001) and 12 hours (81.5 versus 87.1 mm Hg, P=.04) with nifedipine plus diltiazem than nifedipine plus verapamil. Mean nifedipine concentrations were inversely correlated with mean blood pressures. Mean nifedipine area under the curve values were greater with diltiazem than verapamil (1430 versus 1134 ng·h/mL, P=.026), with each greater than nifedipine alone (957 ng·h/mL). Nifedipine plus diltiazem had a greater antihypertensive effect than nifedipine plus verapamil. Diltiazem caused greater increases in nifedipine plasma concentrations than did verapamil. These data suggest that combined calcium channel blockers result in additive antihypertensive effects, perhaps because of a pharmacokinetic interaction.
Calcium channel blockers are used in the treatment of hypertension and other cardiovascular diseases. Although all CCBs inhibit the entry of calcium into excitable cells, they are a heterogeneous group of compounds with distinct structures and pharmacological effects.1 There are three different subclasses of CCBs: benzothiazepines (diltiazem), dihydropyridines (amlodipine, felodipine, isradipine, nicardipine, nifedipine, nimodipine, and nitrendipine), and phenylalkylamines (verapamil). In vitro receptor binding studies with animal tissue models have shown that when diltiazem and a dihydropyridine are used in combination, binding at the dihydropyridine receptor is enhanced.2 3 However, similar studies have shown that the combination of verapamil with a dihydropyridine CCB decreases dihydropyridine receptor binding.3 4 Theoretically, these data suggest that the pharmacological effects of dihydropyridines may be enhanced when combined with diltiazem and diminished when combined with verapamil.
NIF/DIL results in a greater anti-ischemic effect and increased nifedipine Cp values compared with single-agent therapy in patients with angina.5 6 This pharmacokinetic interaction has also been seen in healthy volunteers.7 Clinical trials in hypertension have shown that a dihydropyridine combined with a nondihydropyridine results in an additive antihypertensive response.8 9 10 Despite these promising findings, the use of combined CCBs in hypertension is limited by many unresolved issues and the perception that CCB combinations would not have an additive antihypertensive effect although they have very different pharmacological effects.11
We theorized that combination therapy with a dihydropyridine and nondihydropyridine may offer an additive reduction in BP. This hypothesis has not been prospectively evaluated in a controlled fashion. The purpose of this study was to compare the antihypertensive effects of two combinations of CCBs, NIF/DIL and NIF/VER. In addition, we compared nifedipine pharmacokinetics when given alone and then in combination to determine whether additive antihypertensive effects might be explained by a pharmacokinetic interaction.
Hypertensive subjects were recruited by newspaper advertisement. Subjects were eligible for the study if they were older than 18 years and had a baseline supine DBP between 95 and 115 mm Hg. Permission of the subject's primary care provider was required for enrollment. Premenopausal women were considered for enrollment only if not of childbearing potential. Exclusion criteria were as follows: serious concurrent medical condition (ie, hepatic or liver disease, malnutrition, immunological disorders, alcohol abuse, heart failure or other severe forms of heart disease, including unstable angina, conduction abnormalities, myocardial infarction, or coronary artery bypass graft within the past 3 months), smokers, contraindication to CCB therapy, or use of medications other than antihypertensives.
All subjects were otherwise healthy, as determined by medical history, physical examination, 12-lead electrocardiogram, and routine laboratory evaluation (serum chemistries, complete blood count with differential, and urinalysis). Written informed consent was obtained before subjects were enrolled into the study. The protocol was approved by the institutional review boards of both The University of Illinois Hospitals and Clinics and Rush-Presbyterian–St Luke's Medical Center, Chicago.
This was a randomized, double-blind, three-period crossover trial comparing the steady-state effects of nifedipine monotherapy, NIF/DIL, and NIF/VER in hypertensive adults. The study protocol consisted of a qualification phase and three treatment phases. During the qualification phase, antihypertensive therapy was discontinued in subjects receiving low-dose single-agent therapy. For those who were taking either multiple antihypertensive agents or moderate to high doses of a single agent, medications were tapered over a period of 3 to 7 days and then stopped. BP was then measured weekly to assure that it was less than or equal to 200/115 mm Hg. After a minimum of 2 weeks, when subjects' supine DBP was between 95 and 115 mm Hg, baseline BP and electrocardiographic data were collected between 8 and 9 am. The qualification phase was extended to a maximum of 10 weeks if needed, after which subjects were excluded if the predefined inclusion BP criteria were not met.
During phase 1 of the study, each subject received 30 mg sustained release nifedipine (Adalat CC, lot 3LFC, expiration 10/95, Miles Inc) taken once daily at 8 am for 2 weeks. Subjects were allowed to continue in the study only if supine DBP remained greater than or equal to 90 mm Hg on two or more measurements after nifedipine alone. They were then randomized to receive either 180 mg sustained release diltiazem (Cardizem CD, lot P30128, expiration 10/95, Marion Merrell Dow Inc) or 180 mg sustained release verapamil (Verelan, lot 374-377, expiration 12/95, Lederle Laboratories/Elan Pharma Inc) taken once daily in addition to nifedipine for 2 weeks (phase 2). After a 2-week washout during which they continued nifedipine monotherapy, subjects were crossed-over, based on prior randomization, to receive the other nondihydropyridine for 2 additional weeks (phase 3).
Equal-weight doses of diltiazem and verapamil were chosen because previous studies have shown similar BP reductions in patients with mild to moderate hypertension.12 13 Extended-release drug formulations were selected to enable once-daily dosing and to maximize compliance. Study medications were packaged in single-administration, predated blister packages to enable calculation of compliance by pill count. Diltiazem and verapamil were encased in identical opaque capsule shells to ensure that blinding was maintained throughout the study.
Sample size was calculated prospectively. A difference of 5 mm Hg DBP between combinations was considered clinically significant, and the SD of 6 mm Hg in DBP from a previous analysis was used.14 With a two-tailed test (∝=0.05, β=0.2) 14 subjects were required. However, 16 subjects were sought to assure a statistical power of greater than 80%.
After an overnight fast, subjects reported to the Preventive Medicine Department Clinical Research Center at Rush-Presbyterian–St Luke's Medical Center at 7 am on the last day of phases 1, 2, and 3. At this time, compliance was evaluated, and subjects were questioned for adverse effects related to study medications and were orally dosed at 8 am. BP, heart rate, PR intervals, and nifedipine Cp were measured, in this order, at eight timed intervals after dosing. An investigator who was blinded during phases 2 and 3 measured BPs in duplicate by cuff sphygmomanometry for each subject after the subject had lain supine for 5 minutes and then stood for 2 minutes. Two independent blinded investigators calculated heart rates and PR intervals from the mean of 10 cardiac cycles on the surface electrocardiographic readings using standard methodology. Then, a 7-mL blood sample was collected by venipuncture into heparinized tubes (Vacutainer, Becton Dickinson) for determination of nifedipine Cp values immediately before (0 hour) and 1, 2, 4, 6, 8, 12, and 24 hours after dosing. After collection, the venous blood samples were protected from light and immediately centrifuged at 3000g for 15 minutes. Plasma portions were harvested, transferred to glass tubes, and frozen at −20°C in light-protected containers until assayed. Sample analysis was done in batches by the same blinded technician at a later time. Subjects were fed standard meals at approximately 1, 5, and 9 hours after dosing.
Sample Preparation and Analysis
All nifedipine Cp values were determined by high-performance liquid chromatography with a method developed by the Clinical Research Laboratory, College of Pharmacy, University of Illinois at Chicago, which was modified from the method of Masher and Vergin.15 All sample preparation was carried out in dimmed light to prevent nifedipine degradation. Briefly, 0.5 mL of 1 mol/L NaOH was added to 1 mL plasma, vortexed for 10 seconds, and extracted with 3 mL hexane/methylene chloride (70:30, vol/vol). Samples were placed on a rotating mixer for 7 minutes and then centrifuged for 5 minutes. The organic layer was aspirated and transferred to an amber vial for high-performance liquid chromatographic analysis with an injection volume of 1 mL. The mobile phase was a solution of 90% hexane, 8% 2-propanol, and 2% acetonitrile. Samples were injected into a Waters LC Module 1 column and assayed with the Hewlett-Packard 3359 Automation System. The standard curve was linear over the nifedipine Cp range of 3.75 to 150 ng/mL (r>.999). At quality-control Cp values of 12.6, 37.7, and 101 ng/mL, coefficients of variation were less than or equal to 3.3%, and accuracy was calculated to be between 101% and 102%. A total of 384 plasma samples were analyzed. Two samples (at 0 hour and 24 hours during nifedipine alone in one patient) were below the limit of the assay and were assigned values of 1.88 ng/mL (half the lower limit). Six samples were measured above 150 ng/mL. They were diluted by standard methodology and analyzed with values corrected for dilution.
Noncompartmental analysis was performed with the nifedipine Cp values for each subject during each treatment phase. Maximum Cp (Cmax), time to Cmax (Tmax), and minimum Cp (Cmin) were obtained directly from the data (without interpolation) by visual inspection. Areas under the steady-state Cp curve (AUC0-24) were calculated by the linear trapezoidal rule, and average nifedipine steady-state Cp (Cavg) was calculated from the equation Cavg=AUC0-24/24. Pharmacokinetic data were log-transformed to approximate a normal distribution for the statistical analyses using standard methodology accepted by the Food and Drug Administration (FDA).16
The two one-sided t test bioequivalence criteria recommended by the FDA were used for evaluation of whether a pharmacokinetic interaction was present.17 To accomplish this, 90% confidence intervals (CIs) for two log-transformed AUC0-24 ratios (ie, AUC0-24 during combination therapy to AUC0-24 during nifedipine alone) were calculated. A pharmacokinetic interaction was inferred if these ratios included values outside the accepted range of 0.8 to 1.25.
Statistical differences in supine SBP, supine DBP, heart rate, PR interval, and nifedipine Cp values at each time postdose were evaluated by ANOVA for repeated measures and Scheffé's F procedure for post hoc comparisons. Paired t tests were used for comparisons between the two combination therapies. Linear regression analysis was used for evaluation of the relationships between mean supine BP values and mean nifedipine Cp values. Differences were considered statistically significant at a value of P≤.05. Results are presented as mean±SD.
Twenty subjects were enrolled into the study. Four did not complete the study as originally designed and were excluded from the final analysis; 2 never became hypertensive during the qualification phase; 1 withdrew for personal reasons; and 1 subject's BP was controlled during phase 1. Sixteen subjects (12 men and 4 women) completed all phases of the study. There were 11 blacks, 4 whites, and 1 Asian Indian, with a mean age of 48±8.8 years (range, 31 to 61).
Mean supine SBP and DBP obtained at baseline were 165.8±14.3 and 103.8±5.3 mm Hg, respectively. All three treatments significantly lowered supine SBP (P≤.001) and DBP (P<.02) compared with baseline at all times postdose. Fig 1A⇓ displays mean supine SBP versus time after the dose of study medication during each treatment phase. Both combination therapies resulted in lower mean SBPs compared with nifedipine alone at 0 hour (P=.001), 1 hour (P=.006), and 2 hours (P=.025) postdose. Mean SBPs were statistically lower with NIF/DIL compared with nifedipine alone at 6 hours (P=.04) and 24 hours (P=.04) postdose. Mean supine DBP versus time postdose is shown in Fig 1B⇓. NIF/DIL significantly lowered mean DBP compared with nifedipine alone (P<.05, at all times postdose). Similarly, mean DBPs were lower with NIF/VER compared with nifedipine alone at all times (P<.05), except at 8 hours postdose (P=.11). Mean DBPs were lower with NIF/DIL compared with NIF/VER at 8 hours (77.6±7.4 versus 84.6±6.1 mm Hg, P=.001) and 12 hours (81.5±8.0 versus 87.1±9.7 mm Hg, P=.04) postdose. Mean BP values were highest at 0 hour and 24 hours postdose (trough values). At these times, mean SBPs and DBPs were lowest with NIF/DIL. Changes in standing BPs were similar to those observed with supine BPs.
We evaluated peak antihypertensive effects by calculating the percent maximal reductions in BP values from corresponding baseline values in each subject. The mean percent maximal reductions for nifedipine alone, NIF/DIL, and NIF/VER were 16.1±5.8%, 20.2±6.3%, and 18.9±6.7% in supine SBP and 19.8±4.2%, 27.9±5.0%, and 26.5±5.7% in supine DBP, respectively. We evaluated trough antihypertensive effects similarly by calculating the percent reduction in mean BP values measured at 24 hours postdose from baseline values. The mean percent minimum reductions for nifedipine alone, NIF/DIL, and NIF/VER were 5.3±7.3%, 9.0±6.3%, and 7.7±7.2% in supine SBP and 3.4±4.4%, 9.4±3.8%, and 8.3±4.4% in supine DBP, respectively. These translate into mean peak-to-trough antihypertensive effect ratios (for DBP) of 5.8 for nifedipine alone, 3.1 for NIF/DIL, and 3.2 for NIF/VER.
Mean heart rate and mean PR interval obtained at baseline were 72.2±5.2 beats per minute and 0.168±0.018 seconds, respectively. All observed changes and fluctuations in these parameters throughout the dosing intervals of all treatments were minimal and not statistically significant.
The pharmacokinetic parameters of nifedipine are listed in Table 1⇓. Both combination therapies resulted in increased nifedipine Cp values (Cmax, Cmin, and Cavg) and an overall increase in nifedipine exposure (as indicated by elevated AUC0-24 values). These effects are illustrated in Fig 2⇓, which shows mean nifedipine Cp versus time postdose. Mean nifedipine Cp values tended to be higher with combination therapies compared with nifedipine alone and were highest with NIF/DIL. Nifedipine Cp values with NIF/DIL were greater than those with NIF/VER and nifedipine alone at 0 hour (P=.001) and 12 hours (P=.05). In comparison, nifedipine Cp values with NIF/DIL and NIF/VER were higher than those with nifedipine alone at 1 hour (P=.02), 2 hours (P=.04), and 24 hours (P=.015) but were not different from each other.
The results of the FDA bioequivalence criteria confirmed the observed pharmacokinetic changes. Mean nifedipine AUC0-24 ratios were 1.49 (90% CI=1.31-1.70) with diltiazem and 1.21 (90% CI=1.07-1.37) with verapamil. This suggests that both diltiazem and verapamil interact with nifedipine, resulting in increased drug exposure. Additionally, nifedipine exposures during therapy with NIF/DIL were not bioequivalent to nifedipine exposures during NIF/VER (mean nifedipine AUC0-24 ratio of 1.20; 90% CI=1.03-1.40).
Fig 3⇓ shows scatterplots of the relationships between nifedipine Cp versus supine SBP and versus supine DBP. Linear regression analyses of these relationships are summarized in Table 2⇓. Strong inverse correlations between nifedipine Cp values and both BP values were demonstrated. Slope values were all similar except the nifedipine Cp versus SBP regression line during NIF/DIL therapy. The y intercepts (for both SBP and DBP) of NIF/DIL and NIF/VER were similar and tended to be lower than those of nifedipine alone. This resulted in consistently lower mean SBPs and mean DBPs across the studied nifedipine Cp range with combination therapies compared with nifedipine alone. Because Fig 3A and 3B⇓⇓ plot mean values, a moderate degree of variability was present on both axes.
No period effects were shown. There were no significant differences between 0-hour and 24-hour BP or nifedipine Cp values within treatments (P>.50, paired t test), confirming that apparent steady-state conditions had been met. Pill counts showed compliance to be 99%. None of the subjects returned extra doses of study medication during the last 7 days of any given treatment phase, and 13 of the 16 subjects were 100% compliant.
No serious adverse effects were reported, and no subjects required discontinuation of medication because of side effects. Constipation was the most prevalent side effect, occurring in 8 subjects (50%) receiving NIF/VER. Only 1 subject experienced severe constipation that required corrective therapy (oral laxative and an enema). Three subjects experienced minor leg edema while taking each combination therapy but not during nifedipine monotherapy. Headache was reported in 6 subjects during nifedipine alone, 2 subjects during NIF/DIL, and 5 during NIF/VER.
In designing this study, our primary objective was to compare the antihypertensive effects of two combinations of CCBs directly. Our results consistently show that the addition of diltiazem to nifedipine results in a greater antihypertensive effect than the comparable dose of verapamil given with nifedipine. This observed difference was most pronounced on DBP and was statistically significant at several time points. Although the magnitude of this difference was moderate (maximum difference of 7 mm Hg), these decreases may be clinically important in many hypertensive individuals depending on their degree of BP control.
The degree and duration of antihypertensive response seen with nifedipine alone were poor. Because 30 mg nifedipine is considered a normal starting dose for hypertensive individuals, it is disappointing that only one subject was controlled on this dose. However, after the addition of a nondihydropyridine CCB, BP was markedly reduced. Except for the 0-hour and 24-hour times, mean DBPs were all below 90 mm Hg with each combination therapy.
All three CCBs used in this study were extended-release formulations that allow for once-daily dosing. When clinicians choose an extended-release, once-daily drug formulation, they generally assume that the pharmacological effect is maintained throughout the 24-hour dosing period. This was not observed in our subjects. The peak changes in BP with both combinations were greater than 20% and represent large antihypertensive effects. Unfortunately, these effects were not constant throughout the dosing interval. The largest fluctuations in BPs were seen with nifedipine alone (a 6.3-fold change in peak to trough effect) but were decreased by approximately 50% after the addition of either diltiazem or verapamil. These fluctuations can be explained by the large differences between maximum and minimum nifedipine Cp values over the 24-hour dosing interval. Although the nifedipine formulation used in this study employs an extended-release mechanism for drug delivery, the resultant Cp versus time profiles after dosing of this drug are more characteristic of an immediate-release formulation. This is clearly shown with each combination, for which mean DBP decreased below 80 mm Hg at times of peak effect. Large fluctuations in nifedipine Cp values may also increase the risk of dihydropyridine-related adverse effects with prolonged use.18 19 20 21
Although there are no other clinical trials that directly compare different CCB combinations for hypertension, our findings are similar to other published data. Clinical trials that have evaluated NIF/DIL for angina have demonstrated some degree of additive effect for improving exercise tolerance and ischemic symptoms.5 6 Direct comparisons with these studies are difficult because of differences in markers of drug efficacy, dosing schemes, and the patient population. Toyosaki et al5 used a low-dose fixed combination of 120 mg/d diltiazem and 40 mg/d nifedipine, similar to our dosing with NIF/DIL. As was seen with our NIF/DIL combination, an additive effect was shown. In the treatment of hypertension, the low-dose combination of verapamil with nitrendipine has resulted in a greater BP reduction than twice the dose of each agent given alone.9 A similar additive antihypertensive effect was also seen in our subjects after the addition of verapamil.
With both combination therapies, BPs were reduced in association with both the addition of the nondihydropyridine and increased nifedipine Cp values. NIF/DIL resulted in the largest increases in nifedipine Cp values and the greatest antihypertensive response. The pharmacodynamic relationships (ie, nifedipine Cp versus BP) for each combination therapy were clearly different from that of nifedipine alone. The additive antihypertensive effects provided by the nondihydropyridine caused the NIF/DIL and NIF/VER regression lines (Fig 3⇑) to lie below that of nifedipine alone. If further BP reductions were entirely attributed to elevated nifedipine Cp values, all three regressions would have fallen on the same line. Because this was not seen, we conclude that the BP-lowering effects with NIF/DIL and NIF/VER may have been caused by either a pharmacokinetic interaction, additive effects from the nondihydropyridines, or the combination of these effects.
CCBs are primarily eliminated by hepatic metabolism, with nifedipine being extensively inactivated by hepatic biotransformation.22 Demonstration of an interaction between diltiazem and nifedipine was expected, given the ability of diltiazem to inhibit hepatic oxidative metabolism.23 24 Clinical trials in angina have shown increases of 205%5 and 143% (after dosage adjustment)6 in nifedipine trough Cp values after the addition of diltiazem. Another study showed a 188% increase in the nifedipine area under the curve after 6 days of diltiazem at 180 mg/d,7 although this was observed in normal volunteers after only a single 60-mg dose of nifedipine. Moreover, this normal volunteer study calculated areas under the curve from 0 to 8 hours instead of 0 to infinity, which is standard methodology for single-dose pharmacokinetic studies. Because each of our treatment phases were 2 weeks, we believe our subjects achieved pharmacokinetic steady-state conditions for all CCBs studied. This is the first report to evaluate and detect a clinically significant pharmacokinetic interaction between nifedipine and verapamil. It was perhaps surprising that increases in nifedipine Cp values were larger with diltiazem than with verapamil because verapamil is a more potent inhibitor of oxidative hepatic drug metabolism than diltiazem.25 26
The use of CCB combinations may be exploited for therapeutic benefit. The use of a dihydropyridine CCB in combination with a nondihydropyridine CCB may seem inappropriate because they have traditionally been viewed as part of the same drug class despite their considerable differences. Recently, Materson27 has advocated that this class of drugs be termed calcium antagonists with two subcategories, the dihydropyridines (ie, nifedipine) and the CCBs (ie, diltiazem and verapamil). Recent evidence has linked high doses of nifedipine to increased mortality compared with traditional antihypertensive therapies.19 20 21 Although the risk associated with CCBs has been observed only with immediate-release products, the ability of all the CCBs to prevent morbidity and mortality has been questioned.28 29 Two low-dose extended-release CCBs in combination may be an attractive treatment option in patients requiring moderate to high doses of CCBs to reduce BP to therapeutic goal.
It was not our intention to demonstrate synergy with combination CCB therapy, and our statistical power was not capable of this evaluation. Our use of morning BPs as a surrogate marker of baseline BP status is clearly inferior to a 24-hour ambulatory evaluation, but it is the accepted method of diagnosing hypertension. BPs also may have decreased more if subjects were evaluated after more than 2 weeks of therapy, but we anticipate that any further pharmacological response would have been small. One might wonder how higher doses of nifedipine alone would have compared with the low-dose combinations used in this study. This was not evaluated but could be the focus of future clinical trials. This study assumed equivalence of 180 mg diltiazem to 180 mg verapamil. Although this was based on the medical literature, which has shown that equal-weight immediate-release forms of these agents reduced BP equally,12 13 this may be debated. Our results apply only to the extended-release formulation used in this study. Other once-daily formulations of nifedipine (ie, Procardia XL), diltiazem (ie, Dilacor), and verapamil (ie, Calan SR, Isoptin SR) may minimize pharmacokinetic and pharmacological fluctuations because of their more sustained drug release mechanisms.
In summary, the addition of either diltiazem or verapamil to nifedipine results in an additive antihypertensive effect. This additive effect is greater with diltiazem and is most pronounced near the end of the dosing interval. No antagonistic effects were shown with the concurrent use of nifedipine and verapamil, which does not support in vitro receptor binding studies that have evaluated this combination. Both diltiazem and verapamil cause elevations in nifedipine Cp values, and the magnitude of these elevations is greater with diltiazem. Further BP reductions with combination therapies were greater than the increase in nifedipine Cp values would predict because of additional effects from diltiazem and verapamil. The use of CCB combinations appears to be safe and effective and should be viewed as a potential treatment strategy for hypertension.
Selected Abbreviations and Acronyms
|CCB||=||calcium channel blocker|
|DBP||=||diastolic blood pressure|
|NIF/DIL||=||combined nifedipine and diltiazem|
|NIF/VER||=||combined nifedipine and verapamil|
|SBP||=||systolic blood pressure|
This work was financially supported by a grant from Marion Merrell Dow Inc, and study medication was donated by Miles Inc. We thank Dr Robert Shore for assistance with statistical analyses and insightful comments on the manuscript.
These results were presented at the American College of Clinical Pharmacy 1996 Winter Practice and Research Forum, Monterrey, Calif, February 13, 1996, and the American Society of Hypertension 11th Scientific Meeting, New York, NY, May 18, 1996.
- Received January 10, 1996.
- Revision received February 5, 1996.
- Revision received March 11, 1996.
Toyosaki N, Toyo-Oka T, Natsume T, Katsuki T, Tateishi T, Yaginuma T, Hosoda S. Combination therapy with diltiazem and nifedipine in patients with effort angina pectoris. Circulation. 1988;77:1370-1375.
Frishman W, Charlap S, Kimmel B, Teicher M, Cinnamon J, Allen L, Strom J. Diltiazem, nifedipine, and their combination in patients with stable angina pectoris: effects on angina, exercise tolerance, and the ambulatory electrocardiographic ST segment. Circulation. 1988;77:774-786.
Andreyev N, Bichkov I, Skutelis A, Andreyeva T. Comparison of diltiazem, nitrendipine and their combination for systemic hypertension and stable angina pectoris. J Cardiovasc Pharmacol. 1991;18(suppl 9):s73-s76.
Saseen JJ, Carter BL. Dual calcium channel blocker therapy in the treatment of hypertension. Ann Pharmacother. In press.
Hollifield JW, Heusner JJ, DesChamps M, Gray J. A double-blind, randomized, crossover comparison of equivalent (by weight) oral doses of verapamil and diltiazem in patients with mild to moderate hypertension. J Cardiovasc Pharmacol. 1989;13(suppl 13):s68-s70.
Schuirmann DJ. Treatment of bioequivalence data: log transformation. In: Proceedings of Bio-International 89: Issues in the Evaluation of Bioavailability Data; Toronto, Canada; October 1-4, 1989; 159-161.
Food and Drug Administration. Statistical Procedures for Bioequivalence Studies Using a Standard Two-Treatment Crossover Design. Rockville, Md: Division of Bioequivalence, Office of Generic Drugs, Food and Drug Administration; 1992:1-12.
Myers MG. Dihydropyridine calcium antagonists and the trough:peak ratio: focus on adverse effects. J Hypertens. 1994;12(suppl 8):s73-s77.
Furberg CD, Psaty BM, Meyer JV. Nifedipine dose-related increase in mortality in patients with coronary heart disease. Circulation. 1995;92:1326-1331.
Schlanz KD, Myre SA, Bottorff MB. Pharmacokinetic interactions with calcium channel antagonists, II. Clin Pharmacokinet. 1992;21:448-460.
Materson BJ. Calcium channel blockers: is it time to split the lump? Am J Hypertens. 1995;8:325-329.