Author + information
- Received August 12, 1996
- Revision received January 31, 1997
- Accepted February 21, 1997
- Published online June 1, 1997.
- René R Wenzel, MDA,
- Giuseppe Allegranza, MDA,
- Christian Binggeli, MDA,
- Sidney Shaw, PhDA,
- Peter Weidmann, MDA,
- Thomas F Lüscher, MD, FACCA and
- Georg Nolls, MDA,*
- ↵*Dr. Georg Noll, Cardiology, University Hospital, CH-8091 Zurich, Switzerland.
Objectives. We sought to study the effects of short-acting and long-acting nifedipine on the sympathetic nervous system (SNS), heart rate (HR) and blood pressure (BP) of normotensive subjects under baseline conditions and during SNS stimulation.
Background. Calcium channel antagonists in different pharmacokinetic formulations are widely used in patients with coronary artery disease or hypertension. Short-acting formulations activate the SNS, an action that may be disadvantageous in patients with coronary disease, especially if left ventricular function is impaired. The effects of slow-release formulations on the SNS are unknown.
Methods. We used microneurography to investigate the influence of nifedipine (5 mg; 10 mg; and slow-release [GITS], 60 mg) on muscle sympathetic nerve activity (MSA) and skin sympathetic nerve activity (SSA) in healthy volunteers.
Results. Peak plasma levels after short-acting and slow-release nifedipine were achieved within 60 min and 330 min, respectively. Short-acting (10 mg, n = 10) and slow-release (n = 10) nifedipine, but not placebo, markedly activated MSA and increased plasma norepinephrine; plasma endothelin increased only with slow-release nifedipine. HR increased after short-acting nifedipine, but not after nifedipine GITS. Nifedipine had no effect on SSA (n = 6). Blockade of cardiac sympathetic activity (with esmolol) led to similar decreases in HR with or without nifedipine, whereas parasympatholysis (with atropine) led to similar increases in HR with or without nifedipine. The cold pressor test markedly increased MSA in all treatment groups and further increased MSA beyond the increase induced by nifedipine.
Conclusions. Nifedipine markedly increased MSA, but not SSA, independently of drug release formulation. In contrast, HR increased with short-acting, but not with slow-release, nifedipine. Therefore, nifedipine activates cardiac and peripheral sympathetic nerves differently depending on pharmacokinetics. These effects of nifedipine may be disadvantageous in cardiac patients with increased sympathetic activity or congestive heart failure, or both.
(J Am Coll Cardiol 1997;29:1607–14)
Local or generalized vasoconstriction is an important mechanism in several forms of cardiovascular disease. Augmented vascular tone in the entire circulation accounts for arterial hypertension (). Focal vasoconstriction is an important mechanism in coronary spasm (), contributes to coronary artery disease in general () and is a typical feature of Raynaud’s disease and migraine (). Furthermore, in heart failure, increased peripheral vascular resistance is a secondary phenomenon, but contributes importantly to symptoms and outcome ().
Calcium channel antagonists are effective antihypertensive drugs and exert anti-ischemic effects (). They also exhibit vascular protective properties; improve endothelial function in atherosclerosis and hypertension, at least experimentally (); inhibit proliferation of human coronary artery smooth muscle cells (); and reduce atherosclerosis in hypercholesteremic rabbits (). In the human coronary circulation, dihydropyridine calcium channel antagonists reduce the development of new atherosclerotic lesions ().
Despite these impressive vascular protective effects, clinical trials with calcium channel antagonists have yielded some disappointing results, particularly in patients with coronary artery disease and impaired left ventricular function ([11–16]). A potential mediator of untoward effects of cardiovascular drugs is an activation of the sympathetic nervous system (SNS). Overactivation of the SNS may be detrimental in acute coronary syndromes () and heart failure ([5, 18, 19]). Activation of the SNS may depend not only on the drug used, but also on its pharmacokinetics; that is, the SNS may be stimulated particularly with rapid, but not with delayed, onset of action. Hence, we studied the effects of short-acting and long-acting nifedipine on the SNS, heart rate (HR) and blood pressure (BP) in normotensive subjects under baseline conditions and during SNS stimulation.
1.1 Study Patients.
Studies were performed in healthy volunteers. The subjects were free of cardiovascular or other diseases as assessed by a medical history and a physical examination performed before the study. Written informed consent was obtained from all subjects. The study was approved by the ethical committee of the University Hospital (Inselspital), Bern, Switzerland.
1.2 Experimental Protocol.
All subjects were studied in the morning (9 am) after a light breakfast. After micturition, to avoid any stimulation of sympathetic nerve activity through bladder distension, subjects were asked to resume the supine position. The left or right leg was fixed by a vacuum cushion, and electrocardiographic (ECG) leads, a blood pressure cuff and respiration strain gauge were placed. A catheter (Venflon, Ohmeda, Helsingborg, Sweden) was inserted into a cubital vein. Thirty minutes after puncture of the vein, baseline recordings and blood samplings were performed. When changes in the electrode position occurred, the experiment was discarded.
The following protocols were performed (Fig. 1). In protocol 1, muscle sympathetic nervous activity (MSA) was measured in 10 patients treated with nifedipine 10 mg. In protocol 2, MSA was measured in 6 patients treated with nifedipine, 5 mg. In protocol 3, skin sympathetic activity (SSA) was measured in 5 patients treated with nifedipine, 10 mg. In protocol 4, MSA was measured in 10 patients treated with nifedipine GITS, 60 mg (slow-release form), or placebo. The latter subjects were reexamined on 2 subsequent days, as it is difficult to record MSA for the time period required to achieve maximal nifedipine plasma levels with the GITS formulation. Hence, on day 2, 5 subjects received nifedipine GITS, 60 mg, 180 min before the experiments were started. In protocol 5, 5 subjects of protocol 4 underwent an infusion study with esmolol (300, 1,000 and 2,000 μg/kg body weight) and atropine (1, 3 and 10 μg/kg); the drugs were injected before and after both nifedipine, 10 mg, and nifedipine GITS, 60 mg. These experiments were performed to unmask the sympathetic (esmolol []) or parasympathetic (atropine []) portion of the relative changes in HR (Fig. 1).
All subjects underwent a cold pressor test, with and without drug administration, in which they immersed one hand (up to the wrist) in ice water for 2 min. After the test, the subjects received nifedipine, 5 or 10 mg, nifedipine GITS, 60 mg, or placebo). When estimated peak drug plasma levels were achieved (i.e., 30 min after ingestion of nifedipine, 5 or 10 mg, or 150 min after ingestion of nifedipine GITS or placebo, the test was repeated. In the experiments on day 2, the subjects ingested the drug (nifedipine GITS, 60 mg) 180 min before the test. The test was then repeated 360 min after drug ingestion. If muscle contractions occurred during the maneuvers, causing electromyographic artifacts in the neurogram, the results of the test were discarded.
Blood samples for catecholamine determinations were obtained at baseline and 30 min (150 min for nifedipine GITS) after drug administration. Blood samples for plasma levels of nifedipine were obtained at baseline as well as after 20, 30 and 60 min (nifedipine, 5 and 10 mg) and every 30 min up to 360 min (nifedipine GITS) after drug ingestion.
Multifiber recordings of MSA were obtained from peroneal nerve posterior to the fibular head with tungsten microelectrodes (200-μm shaft diameter: 1 to 5 μm; uninsulated tip). A reference electrode was inserted subcutaneously 1 to 2 cm from the recording electrode. Signal processing was performed as described (). The signal was displayed on an oscilloscope and registered on a thermocoupled printer at a paper speed of 5 mm/s (Graphtec Inc, Yokokawa, Japan). In addition, the nonintegrated signal was recorded on the printer to better identify artifacts. The criteria for acceptance of MSA were selected according to published data (). Neurograms with SSA or mixed SSA and MSA were not accepted. For recordings of SSA, the microelectrode was placed in an area of skin-sensible afference as described (). Accordingly, if there was any evidence of dislocation of the electrode position, the experiment was discarded from the study.
1.4 ECG, Blood Pressure, Respiration.
An ECG was recorded simultaneously on the printer throughout the study. Blood pressure was assessed noninvasively with oscillometric occlusion (Dinamap).
Nifedipine, 5 mg and 10 mg, and slow-release nifedipine (GITS), 60 mg, were given as tablets (Adalat and Adalat CR, Bayer Leverkusen, Germany). For the placebo experiments, identical tablets without active drug were used. Atropine (Sintetica, Mendrisio, Switzerland) and esmolol (Gensia-Opopharma, Zurich, Switzerland) were given as intravenously injected solutions after appropriate dilution with saline solution (0.9%, Inselspital, Bern, Switzerland).
1.6 Plasma Levels of Drug and Hormones.
Nifedipine plasma levels were determined in human plasma by automated electron-capture capillary gas chromatography according to the gas chromatography method (), with modified and validated gas chromatography conditions (ANAWA Laboratories, Zurich).
Plasma catecholamines were determined by high performance liquid chromatography; plasma endothelin was determined by radioimmunoassay (hypertension laboratory, University Hospital, Bern).
1.7 Analysis and Statistics.
MSA was assessed by counting the bursts/min as well as by calculating an arbitrary burst index depending on the amplitude of the bursts. Data are given as mean value ± SEM. We performed an analysis of variance test for treatment (placebo and nifedipine, 5 and 10 mg) and a Bonferroni post-hoc ttest. For nifedipine GITS, 60 mg, a paired ttest was performed to assess the statistical difference versus placebo. For the dose-response curves of atropine and esmolol, the area under the curve was calculated and a paired ttest performed. For the studies with SSA and the cold pressor test experiments, a paired ttest was performed. A p value ≤0.05 was taken for statistical significance.
2.1 Baseline Conditions
2.1.1 MSA and SSA.
Under baseline conditions, MSA was similar in all treatment groups. In the placebo group, MSA (bursts/min and burst index) tended to increase slightly but not significantly during the experiment (Fig. 2, left panel). After administration of nifedipine, 5 mg, MSA remained unchanged (p = NS), but it increased markedly after administration of nifedipine, 10 mg, or nifedipine GITS, 60 mg (p < 0.05 vs. placebo, Fig. 2, left panel). In contrast to MSA, SSA did not change from baseline conditions after nifedipine, 10 mg (12 ± 3 before vs. 14 ± 3 bursts/min) after nifedipine (p = NS).
Thirty minutes after oral administration of nifedipine, 5 or 10 mg, plasma nifedipine levels were 9 ± 5 and 23 ± 7 ng/ml, respectively (p < 0.05 and 0.001). Peak plasma levels were achieved within 60 min after nifedipine, 5 mg (11.6 ± 6 ng/ml, p < 0.05), and within 40 min after nifedipine, 10 mg (29 ± 8 ng/ml, p < 0.001). After administration of nifedipine GITS, 60 mg, nifedipine plasma levels were 3 ± 1 ng/ml within 150 minutes after drug administration; peak plasma levels were reached within 330 min and averaged 12 ± 3 ng/ml.
2.1.2 HR and BP.
At baseline, HR (Table 1) and BP (data not shown) were comparable in all treatment groups. After drug administration, HR increased only in the group treated with nifedipine; 10 mg (p < 0.01), but not in the remaining groups including the group receiving nifedipine GITS, 60 mg (p = NS; Table 1; Fig. 2, right panel).
At baseline, parasympathomimetic doses of atropine (1 and 3 μg/kg) slightly decreased HR, whereas the parasympatholytic dose (10 μg/kg) significantly increased HR (p < 0.01; Fig. 3, left panel). The response to atropine after nifedipine, 10 mg, or nifedipine GITS, 60 mg, was similar (p = NS; Fig. 3, left panel).
At baseline, cardiac beta1-receptor blockade with esmolol induced a significant decrease in HR (p < 0.01; Fig. 3, right panel). After nifedipine, 10 mg, HR tended to decrease more than under baseline conditions (p = NS vs. control; Fig. 3, right panel), whereas after nifedipine GITS, 60 mg, the decrease was similar to baseline conditions (p = NS vs. control; Fig. 3, right panel).
Systolic BP of these normotensive subjects did not change after drug administration in either group (p = NS, data not shown); diastolic blood pressure decreased slightly after nifedipine, 10 mg (−5 ± 2 mm Hg, p < 0.01), whereas it slightly increased 150 min after nifedipine GITS, 60 mg (+5 ± 3 mm Hg, p < 0.05).
2.1.3 Plasma Catecholamines.
Plasma norepinephrine markedly increased 30 min after nifedipine, 10 mg (from 160 ± 21 to 238 ± 26 pg/ml [n = 4], p < 0.05), whereas the increase in plasma epinephrine was not statistically significant (from 19 ± 7 to 37 ± 11 pg/ml [n = 4], p = NS vs. baseline). After nifedipine GITS, 60 mg, plasma norepinephrine increased 150 min after drug intake (from 147 ± 33 to 214 ± 45 pg/ml [n = 10], p < 0.05), further increased 300 min after nifedipine GITS, 60 mg, on day 2 (264 ± 112 pg/ml [n = 5], p = NS vs. baseline) and decreased again after 360 min (172 ± 21 pg/ml [n = 5], p = NS vs. baseline). Plasma epinephrine values tended to increase after both nifedipine, 10 mg, and nifedipine GITS, 60 mg (nifedipine, 10 mg, from 19 ± 7 to 37 ± 11 pg/ml [n = 4], p = NS; nifedipine GITS, 60 mg, from 51 ± 10 to 64 ± 25 pg/ml [n = 10], p = NS vs. baseline; peak value 330 min after drug intake 85 ± 42 pg/ml [n = 5], p = NS vs. baseline).
2.1.4 Plasma Endothelin.
At baseline, plasma endothelin was 2.1 ± 0.1 pg/ml. After nifedipine, 10 mg, it tended to increase (2.2 ± 0.1 pg/ml, p = NS), whereas it increased markedly 6 h after nifedipine GITS, 60 mg (2.9 ± 0.4 pg/ml, p < 0.05 [n = 5]).
2.2 Cold Pressor Test
2.2.1 MSA and SSA.
Under control conditions, the cold pressor test led to the strongest increase in MSA in all treatment groups (p < 0.01 to p < 0.05; Fig. 4). Although MSA increased (p < 0.05) after both nifedipine, 10 mg, and nifedipine GITS, 60 mg, the cold pressor test increased MSA further and in an additive fashion. However, the difference between the increase of MSA with the cold pressor test and MSA under baseline conditions was significant only in the placebo group (p = 0.05; Fig. 4) and in the group treated with nifedipine, 10 mg (p < 0.05 vs. control; Fig. 4).
Under control conditions, SSA increased significantly with the cold pressor test (+10 ± 2 bursts/min, p < 0.01). However, after nifedipine, 10 mg, the increase in SSA during the cold pressor test was less pronounced (+6 ± 2 bursts/min, p < 0.05 vs. control).
2.2.2 HR and BP.
HR tended to increase with the cold pressor test under control conditions (p = NS; Table 1), whereas both systolic and diastolic BP markedly increased with the test (+22 ± 2/+20 ± 1 mm Hg [systolic/diastolic], p < 0.01 to p < 0.05). HR during the cold pressor test was significantly higher after nifedipine, 10 mg (p < 0.001 vs. control), but not after nifedipine GITS, 60 mg (p = NS; Table 1). Both systolic and diastolic BP were similar during the cold pressor test before and after nifedipine, 10 mg, and nifedipine GITS, 60 mg (p = NS; data not shown).
This study is the first to demonstrate with direct measurement of sympathetic nerve activity in the peroneal nerve in humans that nifedipine independent of the drug release formulation, is a strong stimulator of peripheral sympathetic nerve traffic. However, with the short-acting compound, both cardiac sympathetic activity (i.e., HR) and MSA increased, but with the slow-release preparation, HR remained unchanged although MSA increased to a similar degree. There was no increase in cardiac parasympathetic tone after nifedipine GITS. Nifedipine did not affect SSA.
3.1 Role of SNS in Cardiovascular Disease.
The SNS is an important regulator of the circulation whose activity is determined by psychologic, neuronal and humoral factors (). The effects of cardiovascular drugs on the SNS have been studied little and are most commonly assessed by measuring plasma catecholamines. However, plasma catecholamine determinations provide only an indirect measure of sympathetic nerve activity because only the overflow of the adrenergic neurotransmitter from the synaptic cleft is measured. Furthermore, plasma catecholamine levels reflect the activity not only of adrenergic neurons, but also of the adrenal medulla. Finally, most methods of measuring plasma catecholamines are prone to considerable variation. In contrast, microneurography directly records sympathetic nerve traffic ([23, 27]). Because the signals can be obtained on-line, the method allows recordings of small and short-lasting changes and their time course during stimulatory maneuvers. The latter property of microneurography allows characterization of changes in sympathetic nerve activity during administration of a cardiovascular drug and analysis of the importance of pharmacokinetic properties of a given preparation under these conditions.
3.2 Effects of Nifedipine on SNS.
Our study confirms previous reports of the effects of nifedipine, 10 mg, in particular the increase in HR occurring after application of a short-acting dihydropyridine in humans ([28–31]). Furthermore, our microneurographic data show that under these conditions a parallel increase in MSA traffic as well as an increase in norepinephrine plasma levels also occurs. The degree of activation of MSA was impressive and comparable to that obtained with a cold pressor test, one of the strongest activators of MSA (). Even more surprising than the degree of activation of MSA was the fact that in subjects receiving nifedipine the system could be further stimulated during the cold pressor test to a degree similar to that without the drug and—especially with nifedipine, 10 mg—in addition to an already increased baseline activity. This finding indicates that with short-acting dihydropyridines, the peripheral and cardiac portions of the SNS are highly activated and remain as responsive to these stimulatory maneuvers as they are in untreated subjects.
SSA was not influenced by nifedipine, 10 mg, indicating that this part of the SNS is regulated by different mechanisms. This observation underscores the concepts that sympathetic outflow is not uniformly regulated but specifically targeted to different organs and tissues and that nifedipine acts on the part of the SNS regulated by the baroreflex.
3.3 Effects of Slow-Release Nifedipine (GITS) on Cardiac SNS.
Gastrointestinal therapeutic systems (GITS) have been developed to change the pharmacokinetic properties of short-acting molecules. The pharmacokinetics of nifedipine GITS, 60 mg, is very different from that of the original formulation ([33, 34]). Indeed, peak plasma levels were reached within 40 to 60 min after oral administration of nifedipine, 5 or 10 mg, whereas they occurred at 330 min after administration of nifedipine GITS, 60 mg. It has been suspected that activation of the SNS might depend on the pharmacokinetics of the drugs, that is, on the speed of unloading of the carotid baroreceptors. On the bases of these short-term experiments, it appears that—in contrast to short-acting nifedipine—nifedipine GITS, 60 mg, results in no change in cardiac sympathetic outflow. Indeed, HR did not change significantly with nifedipine GITS, 60 mg, whereas a marked increase was noted with nifedipine, 10 mg. Furthermore, increases in HR over control values in response to parasympatholytic dosages of atropine were similar after both short-acting and slow-release nifedipine; therefore, the missing increase in HR after slow-release nifedipine is not due to increased parasympathetic tone. In contrast, the decrease in HR after cardiac sympatholysis with esmolol was similar with and without nifedipine GITS, whereas it was slightly higher after short-acting nifedipine; this finding demonstrates, that after slow-release nifedipine as opposed to short-acting nifedipine, cardiac sympathetic activation does not occur despite a marked increase in peripheral sympathetic tone.
3.4 Effects of Slow-Release Nifedipine (GITS) on Peripheral SNS.
In contrast to cardiac sympathetic outflow, MSA showed a completely different pattern. Under both conditions, that is, with administration of the short-acting and the long-acting forms of nifedipine—MSA increased quite dramatically and to a similar level, although peak nifedipine plasma levels were actually lower with nifedipine GITS, 60 mg, than with nifedipine, 10 mg. The only difference between the two preparations was the more subtle onset of MSA with the slow-release form. Indirect measurements of sympathetic activity measuring forearm blood flow and forearm norepinephrine spillover () have suggested an increase of sympathetic activity despite a lack of heart rate increase during both short- and long-term treatment with the calcium channel antagonist felodipine in hypertensive patients. Hence, our study directly demonstrates that the SNS is not uniformly activated and that cardiac sympathetic activity and MSA are differentially regulated. It appears that with a rapid onset of vasodilation (induced in this instance by calcium channel antagonists), the cardiovascular system attempts to compensate for these changes by a generalized increase in sympathetic nerve activity leading to increased HR and peripheral vascular resistance. This may explain why, despite rather high drug levels of nifedipine, BP changed only very slightly in these normotensive subjects and no hypotension occurred. With a more subtle onset of vasodilation with a slow-release preparation of nifedipine, it appears to be sufficient to selectively increase MSA while leaving HR unchanged. Possibly, baroreflex sensitivity controlling HR adapts rapidly with the slow-release formulation, whereas peripheral sympathetic activity is still activated ().
Plasma norepinephrine and plasma endothelin levels increased after nifedipine GITS, 60 mg. The fact that plasma endothelin levels increased only after slow-release nifedipine indicates that endothelin is newly synthesized, which takes several hours (). Catecholamines such as epinephrine, which stimulate the synthesis of endothelin, may be potential mediators (). The increases in MSA, plasma norepinephrine and plasma endothelin explain why, despite a missing increase in HR, BP did not decrease after nifedipine GITS.
3.5 Potential Therapeutic Implications.
These findings may have important therapeutic implications. Indeed, the fact that plasma norepinephrine levels are a very strong predictor of mortality in patients with heart failure () suggests that activation of the SNS is detrimental at least in these patients, but possibly also in other patient groups, such as those with hypertension (). Indeed, it has been well documented () that short-acting calcium channel antagonists should not be used in unstable clinical situations such as hypertensive crisis or myocardial ischemia. Whether slow-release formulations of nifedipine are safe in hypertensive patients with or without coronary artery disease is under investigation.
Although the data of the present study cannot necessarily be extrapolated to patients with heart failure or hypertension, as both baroreflex control and baseline sympathetic activity are altered under these conditions, assessment of the effects of cardiovascular drugs on the SNS may be an important step in the evaluation of these patients. Indeed, angiotensin-converting enzyme inhibitors such as captopril do slightly lower MSA under the same experimental conditions applied in the present study (). Similarly, it is likely that all drugs activating the SNS are detrimental in patients with an acute coronary syndrome (). However, the differential regulation of peripheral and cardiac sympathetic outflow, particularly with slow-release forms of dihydropyridines but potentially also with other drugs, suggests that the effects of short-acting drugs and slow-release forms may differ considerably in these particular clinical syndromes.
3.6 Potential Mechanisms.
Examination of the mechanism of activation of the SNS by dihydropyridine calcium channel antagonists is beyond the scope of this study. Most likely, however, the vasodilation leads to an activation of baroreflex activity and, in turn, to a stimulation of sympathetic outflow. A direct stimulation of the SNS through central mechanisms by dihydropyridines is unlikely. Indeed, there are no experimental data supporting such an interpretation, although these drugs are able to reach tissues of the central nervous system ().
The lack of an increase in HR demonstrates a differential regulation of cardiac and peripheral sympathetic activity. Therefore, changes in HR do not necessarily reflect changes in the peripheral SNS and cannot be used as a surrogate for the measurement of the activity of this important regulatory mechanism when vasoactive drugs are tested.
☆ This study was supported by grants from the Swiss National Foundation, Bern (32-32 655.91 and 32-35 591.92 [SCORE]), and grants from Bayer, Leverkusen, Germany and the Roche Research Foundation, Basel, Switzerland. Dr. Wenzel is recipient of a stipend by the German Research Association, Bonn, Germany (Deutsche Forschungsgemeinschaft, No. WE 1772/1-1).
- blood pressure
- electrocardiogram, electrocardiographic
- heart rate
- gastrointestinal therapeutic system (slow-release formulation of nifedipine)
- muscle sympathetic nerve activity
- sympathetic nervous system
- skin sympathetic nerve activity
- Received August 12, 1996.
- Revision received January 31, 1997.
- Accepted February 21, 1997.
- The American College of Cardiology
- Lund-Johansen P
- Nayler WG,
- Szeto J
- Yang Z,
- Noll G,
- Lüscher TF
- ↵Lichtlen PR, Hugenholtz PG, Rafflenbeul W, Hecker H, Jost S, Deckers JW. Retardation of angiographic progression of coronary artery disease by nifedipine: results of the International Nifedipine Trial on Antiatherosclerotic Therapy (INTACT). INTACT Group Investigators. Lancet 1990;335:1109–13.
- ↵HINT: early treatment of unstable angina in the coronary care unit: a randomised, double blind, placebo controlled comparison of recurrent ischaemia in patients treated with nifedipine or metoprolol or both: report of The Holland Interuniversity Nifedipine/Metoprolol Trial (HINT) Research Group. Br Heart J 1986;56:400–13.
- Held PH,
- Yusuf S,
- Furberg CD
- Secondary prevention with verapamil after myocardial infarction: the Danish Study Group on Verapamil in Myocardial Infarction. Am J Cardiol 1990;66:33I–40I.
- Furberg C,
- Psaty BM,
- Meyer JV
- McCance AJ,
- Forfar JC
- Swedberg K,
- Eneroth P,
- Kjekshus J,
- Wilhelmsen L
- Leimbach WJ,
- Wallin BG,
- Victor RG,
- Aylward PE,
- Sundlof G,
- Mark AL
- Noll G,
- Wenzel RR,
- Schneider M,
- et al.
- Reid JL
- Ferguson DW,
- Hayes DW
- Kiowski W,
- Erne P,
- Bertel O,
- Bolli P,
- Bühler F
- ↵Noll G, Wenzel RR, De Marchi S, Lüscher TF. Differential modulation of the sympathetic nervous system by captopril and nitroglycerin. Circulation. In press.
- Kerckhoff W,
- Drewes LR