Calcium channel blockers

This chapter is probably relevant to Section G8(iii)  of the 2017 CICM Primary Syllabus, which asks the exam candidate to "understand the pharmacology of anti-hypertensive drugs".  Calcium channel blockers, like beta-blockers, straddle the antihypertensive and antiarrhythmic functional classifications. So as not to duplicate content, insulting the reader with repetitive material, the antiarrhythmic properties of these drugs will be carelessly disregarded in this chapter. The main focus will be on common properties of these drugs, and in particular on producing some sort of structure which might help the candidates answer future questions, as lamenting a lack of structure seems to have been the most common college examiner comment. 

Historically, the SAQs on calcium channel blockers have not asked anybody to compare between the properties of drugs from the same class, but rather to "describe the pharmacology"

• Question 8 from the second paper of 2017 (calcium channel blockers, nimodipine)
• Question 2 from the first paper of 2014  (calcium channel blockers, verapamil)
• Question 17 from the second paper of 2011 (calcium channel blockers)

In summary:

Godfraind (2017) is probably the best reference for structure/function relationships, which makes some sense given that Theophile Godfraind is basically the father of modern calcium channel blockers, working in the field since 1960s. He also published an excellent thing in 1994 which is still the best resource for a discussion of calcium channel blocker cardioselectivity, and which happens to be free from Springer.  For pharmacokinetics, one can't go past Kelly & O'Malley (1992), and for the mechanism of action and clinical effects Abernethy & Schwartz (1999) is the most concise paper, though unfortunately paywalled by NEJM.  The trainees are reminded that the detailed scrutiny of these references in the pursuit of a comprehensive knowledge of calcium channel blockers should not be viewed as an effective use of their final pre-exam hours, as the grey box above probably already represents something well in excess of the expected minimum knowledge.

Available calcium channel blockers

An excellent peer-reviewed account of how these drugs were discovered and developed can be found in Drapak et al (2017) and Godfraind (2017), made more authentic by the personal involvement of the latter. In short, we have this class of drugs because back in the early 1960's a whole series of Big Pharma giants threw money at the problem of finding new any-anginal agents. Specifically, they were looking for coronary vasodilators, armed with some new research frameworks outlined by Arunlakshana and Schild (1959). Ultimately, as the result of this broad screening of multiple molecules,  Bayer gave us dihydropyridines, Tanabe gave us diltiazem, Knoll produced verapamil, and Janssen Pharmaceutica produced a whole series of vasoactive substances which went nowhere (lidoflazine, cinnarizine, flunarizine).

Calcium Channel Blockers

Drug	Year of first availability
Nifedipine	1981
Verapamil	1981
Diltiazem	1982
Nimodipine	1985
Amlodipine	1990
Lercanidipine	1997
Clevidipine	2007

Anyway. The history of calcium channel blockers is tied closely to the history of calcium channel research, and verapamil already existed in some nameless form even in the early 1960s, as a substance which was known to mimic the effects of severe hypocalcemia on myocyte cultures. From the first available agents (verapamil and nifedipine), the main improvements in chemistry have focused on producing drugs which were longer-acting, allowing for once-daily dosing. 

Several other agents are on the market (felodipine, nicardipine, isradipine, etc), but their market share in Australia is smaller, and moreover it was felt that all the dihydropyridines are so very similar that from an educator's perspective it would be pointless to list them all. As such, only drugs with specific applications (nimodipine) or curious pharmacokinetic weirdness (clevidipine) will be singled out for discussion here. Also, amlodipine, because according to Eliott & Ram (2011) it accounts for about 70% of calcium channel blocker sales. There is not much to know about the pharmaceutics of these drugs, except that all of them (with the exception of diltiazem and nifedipine) are sold as racemic mixtures, where one of the stereoisomers has absolutely no activity. 

Routes of administration and GI absorption

Diltiazem, verapamil, nimodipine have alternative IV formulations, and clevidipine is the only one which is not available as a tablet. Their other pharmacokinetic parameters, listed below, come from Kelly & O'Malley (1992) where Tabel II also lists the references from where these numbers were derived. Clevidipine was not available in 1992, so those data were extracted from Nordlander et al (2004). Everywhere else, (potentially broken) links point to the relevant resource;.

Bioavailability of Calcium Channel Blockers

Drug	Available routes	Bioavailability	pKa
Nifedipine	Oral	45%	3.93
Verapamil	Oral or IV	24%	8.73
Diltiazem	Oral or IV	39%	7.5
Nimodipine	Oral or IV	11.6%	5.4
Amlodipine	Oral	64%	8.6
Lercanidipine	Oral	10%	9.3
Clevidipine	IV only	0%	5.3

Solubility and protein binding

For the most, these drugs are highly protein-bound, and extremely lipophilic. From Kelly & O'Malley (1992), Pang et al (1984) and Uceda et al (1995):

Solubility and Protein Binding of Calcium Channel Blockers

Drug	Protein binding	Lipid solubility (octanol:water partition coefficient)
Nifedipine	96-98%	Basically insoluble in water
Verapamil	84-91%	67.0
Diltiazem	80-86%	2.7
Nimodipine	98%	3.8
Amlodipine	95.5%	3.0
Lercanidipine	98%	6.0
Clevidipine	99.5%	Basically insoluble in water

These data have implications for their clearance in severe toxicity. Dialysis is unlikely to get rid of enough molecules to make much of a difference over a reasonable timeframe. However, the high lipid solubility of some of these drugs (eg. amlodipine) has prompted some authors to recommend a lipid emulsion as an antidote. Sebe et al (2015) reported on three years of experience (15 patients) and described the results as positive, in the face of a 20% adverse event rate.

The lipid solubility of these drugs also has some non-toxicological relevance. Zweiten (1998) discussed that, with the increased distribution of these substances into the lipid bilayer, membrane depots should theoretically form, which should therefore lead to a longer duration of action and slower onset of activity. This is a good segue to...

Half-life and duration of effect

The time to peak effect and the elimination half-life are mainly from Abernethy & Schwartz (1999):

Onset of Effect and Half-Life of Calcium Channel Blockers

Drug	Time to peak effect	Elimination half-life
Nifedipine	0.5 hrs	2 hrs
Verapamil	0.5-1.0 hrs	4.5-12 hrs
Diltiazem	0.5-1.5 hrs	2-5 hrs
Nimodipine	1 hr	1-2 hrs
Amlodipine	6-12 hrs	30-50 hrs
Lercanidipine	1-3 hrs	2-5 hrs
Clevidipine	seconds	1.8 min

As you can see from this table, the elimination (plasma) half-lives of these agents are completely unrelated to the duration of their clinical effect. For example, lercanidipine has a relatively short half-life, but because of its high lipid solubility it forms a lipid depot and associates with its target for longer, which means it only needs to be taken once daily.

Out of these drugs, clevidipine is clearly an isolated abnormality, owing to its extremely rapid hydrolysis by plasma esterases. Which brings us to the topic of metabolism:

Distribution, metabolism and elimination

Kelly & O'Malley (1992) have this excellent table which summarises the pharmacokinetics of then-available drugs. 

Distribution and Metabolism of Calcium Channel Blockers

Drug	Volume of Distribution	Metabolism
Nifedipine	13 L/kg	Mainly hepatic, by CYP3A4
Verapamil	3.8 L/kg	Mainly hepatic, by CYP3A4 (which it inhibits)
Diltiazem	5.3 L/kg	Mainly hepatic, by CYP3A4 (which it inhibits)
Nimodipine	1.7 L/kg	Mainly hepatic, by CYP3A4
Amlodipine	21.4 L/kg	Mainly hepatic, by CYP3A4
Lercanidipine	2.5 L/kg	Mainly hepatic, by CYP3A4
Clevidipine	0.56 L/kg	Plasma esterases

CYP 3A4 is important to mention, because it is the pharmacokinetic workhorse of the liver. Fully 50% of routinely used xenobiotics end up being metabolised by this enzyme isoform. It is important to point out that verapamil and diltiazem also act as inhibitors for CYP 3A4,  in addition to relying on it for their clearance. 

Why is is this miniscule detail being mentioned here? Well. Question 8 from the second paper of 2017, which was all about nimodipine, specifically asked for an answer "including important drug interactions". Thus, following from the above, the metabolic clearance of nimodipine would be increased by phenytoin, rifampicin and corticosteroids, and decreased by diltiazem, verapamil, erythromycin, and grapefruit.

Chemical structure and structure-function relationships

There's a couple of ways to classify CCBs. One way would be to draw a functional divide between agents which are mainly cardioselective and agents which are mainly vasodilatory. That way, you would end up with a dihydropyridine group and a non-dihydropyridine group.  This is what many textbooks tend to do. Judging by the college answer to Question 17 from the second paper of 2011, the examiners wanted us to subdivide things even further, and separate the cardioselective group into phenylalkylamines and benzothiazepines:

• Phenylalkylamines:
  • Verapamil
• Benzothiazepines:
  • Diltiazem
• 1,4-dihydropyridines:
  • Nifedipine
  • Nimodipine
  • Amlodipine
  • Lercanidipine
  • Clevidipine

For some reason, the examiners had offered numbered Types instead of chemical classes ("Type I : 
Phenylalkylamines eg verapamil, Type II : Dihydropyridines eg nimodipine and Type III : 
Benzothiazepines eg diltiazem"). This seems to be an alternative naming strategy which you only see in some of the older papers (eg.  Singh, 1986), and it will not be used here because it fails as nomenclature (i.e. numbered class names describe nothing useful) and because it might breed confusion with the already confusing number-based classification of antiarrhythmic agents. 

Anyway: structure and function. The best resource for this was a 1991 article by David Triggle, which is fortunately available in full text. In short, each of the abovementioned classes is chemically distinct from the other, which makes them somewhat difficult to discuss. Dihydropyridines are relatively easy, because these are a large group of molecules which all share a common structure. There's a phenyl ring with a variable substitution position, and two groups at the C3 and C5 positions. The latter seem important - the substitution of NO₂ for CH₃ here turns isradipine into a calcium channel agonist.

Things are somewhat more complex for verapamil and diltiazem, as their structures are totally different. Verapamil and its close chemical relative gallopamil are phenylalkylamine derivatives and diltiazem is a benzothiazepine.  These drugs have completely different molecular shapes, which causes them to interact with their molecular target in a completely different way and this accounts for their cardioselectivity.

Which brings us to...

The molecular drug target of calcium channel blockers

One can summarise this whole section by saying that calcium channels are a complex pore-forming transmembrane protein, and all the commercially available calcium channel blockers interfere with its activity in one of three distinct ways. It is possible to go completely mad with detail here, but all you really need to know is summarised in the second paragraph of Abernethy & Schwartz (1999). In short:

• L-type calcium channels are complex transmembrane proteins which are made up of four repeating sections (referred to as "motifs") each made up of six subunits. 
• All calcium antagonists bind to the α_1c subunit of the L-type calcium channel
  • Dihydropyridines  bind to transmembrane segment 6 of motif III and IV
  • Phenylalkylamines bind to transmembrane segment 6 of motif IV
  • Benzothiazepines bind to the cytoplasmic bridge between motif III and otif IV (IVS), and the (nifedipine-like) calcium antagonists (IVS6) (Fig. 1).

Different classes of calcium channel blockers interact with their targets in slightly different ways, which explains their selectivity and different clinical effect:

• 1,4-dihydropyridines, eg. nifedipine, are allosteric modulators of L-type calcium channels. By interacting with its external lipid-facing surface at the junction of two subunits, these drugs force an allosteric change in the selectivity filter of the pore, essentially blocking it with the next available calcium ion (Tang et al, 2016).
  • This property seems to be partly dependent on the resting membrane potential of the cell which bears the calcium channels. Soldatov et al (1995) found that isradipine had 8.6 times less inhibitory potency when the membrane potential was dropped from -40 to -90 mV.  This is a drug-specific effect, and is thought to account for some of the apparent selectivity of some dihydropyridine drugs for different vascular territories. For example, cerebral vessels are more sensitive to the effects of nimodipine because their resting membrane potential is lower (Morel & Goldfraind, 1989).
  • Also, for the same reason, these drugs are slow to interact with receptors in depolarised tissues, which means that they are less likely to interact with L-type calcium channels in the myocardium (Godfraind, 2017).
  • On top of that, there are numerous isoforms of the L-type calcium channel, and there are differences in drug affinity for the cardiac isoform vs. the vascular isoform of the α_1C-a subunit.
  • Lastly, calcium channels are not the only channels with which these drugs can interact. At a high enough dose, they will also block sodium channels. Yatani et al (1988) determined that nitrendipine achieved voltage-gated sodium channel blockade at extracellular fluid concentrations around 5-10 times the normal therapeutic range.
• Phenylalkylamines and benzothiazepines, eg. verapamil and diltiazem, are "proper" pore blockers; which is to say they get inside the channel pore and physically prevent ion traffic. They also do not seem to care overmuch which channel isoform they have bound to, or what the resting membrane potential is. Morel et al (1998) determined that verapamil produced the same level of block in both cardiac and vascular isoforms of the L-type calcium channel, irrespective of whether the membrane potential was -50 mV or -100 mV. 

Thus, the potency of the vasodilator effect of these drugs tends to vary between tissues, and between drugs. Most authors (eg. Abernethy & Schwartz, 1999) seem to think that nifedipine is the most potent vasodilator, and verapamil and diltiazem are the least potent. Between verapamil and diltiazem, verapamil is said to have the greater negative inotropic and chronotropic potency.

Mechanism of action

To briefly recap the physiology of L-type calcium channels:

• L-type calcium channels are voltage-gated calcium channels, which open (slowly) at a threshold membrane potential of around -30 mV, and conduct a calcium current into the cell
  • In the cardiac muscle, these account for the Phase 2 plateau of the normal cardiac action potential.
  • In cardiac pacemaker tissue, these channels are responsible for Phase 0 of the pacemaker potential
  • In vascular smooth muscle, these channels are responsible for regulating intracellular calcium concentration, which determines the smooth muscle tone (Hughes et al, 1995).
• In each case, these channels open to allow an influx of calcium, which then facilitates the release of even more calcium from the ER and SR of muscle cells.
• The increased intracellular availability of calcium leads to increased myocyte contraction because of increased calcium-troponin-myosin interaction

Thus, calcium channel blockers...

• Relax vascular smooth muscle, thereby decreasing peripheral vascular resistance and afterload (in the case of all CCBs, but especially the dihydropyridines)
• Decrease cardiac contractility and decrease heart rate, thereby decreasing myocardial oxygen demand (in the case of verapamil and diltiazem)

The effect on blood pressure for these drugs is said to be antihypertensive, and the CICM trainee should certainly be instructed to enter this into their written exam answer. However, as in the corresponding section in the beta-blocker chapter, the effect of this vasodilation on the circulatory system is moderated by compensatory mechanisms.  Consider: when calcium channel blockers produce the relaxation of vascular smooth muscle, the resulting drop in blood pressure should immediately stimulate the cardiac reflexes that are in charge of maintaining a stable blood pressure, increasing the sympathetic stimulus for cardiac contractility and heart rate, an effect which should offset most of the antihypertensive effect. 

What is the magnitude of these compensatory effects? Logic dictates that they should theoretically be able to compensate for 100% of the change, rendering calcium channel blockers completely ineffective; and obviously this cannot be the case because it would severely hamper all efforts at marketing these drugs as antihypertensive agents. The reality appears to be some compromise, where the sympathetic activation meets the vasodilation half-way, and the blood pressure still drops, albeit not as much as it would if the baroreflexes were out of the equation.  When Kailasam et al (1995) measured the sympathetic reflex responses to felodipine, they observed a drop in blood pressure (MAP fell by about 10mmHg) and a compensatory increase in heart rate (HR increase from 60something to 70something), accompanied by a 20% increase in circulating plasma catecholamines. This effect diminished with time, and was not sustained chronically, suggesting some level of tired complacency on the part of the baroreflexes (the reflex tachycardia was basically absent by the fourth week of therapy).

Additionally it appears there are some class-dependent differences in the extent of this effect. Verapamil did not produce any such compensatory baroreceptor effect, and in fact the circulating levels of catecholamines had decreased by about 20% instead. The authors go on to point out that verapamil, apart from its cardiodepressant effects (which directly antagonise baroreceptor responses) also has a host of other sneaky effects. For example, it can deplete catecholamine vesicular stores,  directly bind to α-receptors, suppress presynaptic noradrenaline release directly by a calcium channel effect, and penetrate the blood brain barrier, going god-only-knows-what to the central control of sympathetic function. In short, for dihydropyridine calcium channel blockers the antihypertensive effect is initially resisted by baroreceptor responses which slowly adapt with chronic use, whereas for the non-dihydropyridine CCBs, numerous non-class effects interfere with sympathetic reflex responses.

Clinical effects and indicated uses

Abernethy & Schwartz (1999) are again probably the best resource for a detailed overview of this topic. For the ICU trainee, their elaborate digressions on various outpatient antihypertensive therapy trials are probably irrelevant. 

Desirable effects:

• Systemic vasodilator effects:
  • Acute control of severe hypertension (clevidipine)
  • Control of chronic hypertension (dihydropyridines)
  • Afterload reduction without negative inotropic effects  (dihydropyridines)
• Regional vasodilator effects:
  • Prevention of cerebral vasospasm (nimodipine)
• Heart rate control (verapamil and diltiazem):
  • Where beta-blockers are contraindicated, eg. where peripheral vascular disease, uncontrolled diabetes or severe bronchodilator-dependent asthma make you think twice about the use of these agents)
• Antiarrhythmic effects
  • In the same way, where beta-blockers are contraindicated, verapamil or diltiazem can be used to control rate, or maintain sinus rhythm, in supraventricular arrhythmias

Adverse effects:

• Extension of the beneficial effects:
  • Negative inotropy
  • Bradycardia, hypotension
• Unrelated to blood pressure and heart rate
  • Gingival hyperplasia
  • Peripheral oedema
  • Headache 
  • Constipation (with verapamil and diltiazem)
• Unpleasant interactions:
  • With beta-blockers (additive effect, giving rise to heart block or asystole)
  • With other CYP3A4 inhibitors or activators - eg. phenytoin and rifampicin can increase the clearance of these drugs, whereas in turn verapamil and dialtiazem can inhibit the clearance of other CYP3A4 substrates.