Beta blockers vs. calcium channel blockers

National Clinical Guidelines Centre (UK)

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Clinical Guidelines Centre (UK). Stable Angina: Methods, Evidence & Guidance [Internet]. London: Royal College of Physicians (UK); 2011 Jul. (NICE Clinical Guidelines, No. 126.)

See 2016 Partial Update

Update information August 2016: NICE added a footnote to recommendations 1.4.11 and 1.4.12 covering the new advice from the Medicines and Healthcare products Regulatory Agency (MHRA) about safety concerns related to ivabradine (June 2014 and December 2014) and nicorandil (January 2016).

Update information August 2016: NICE added a footnote to recommendations 1.4.11 and 1.4.12 covering the new advice from the Medicines and Healthcare products Regulatory Agency (MHRA) about safety concerns related to ivabradine (June 2014 and December 2014) and nicorandil (January 2016).

Stable Angina: Methods, Evidence & Guidance [Internet].

Show details

Contents

< Prev Next >

7Beta blockers vs. calcium channel blockers

7.1. Introduction

Anti-anginal drugs prevent attacks of angina by decreasing myocardial oxygen consumption (by lowering heart rate, blood pressure, myocardial loading, or myocardial contractility) and/or by increasing myocardial oxygen supply (by increasing coronary blood flow).

Evidence that monotherapy with single anti-anginal agents prevents attacks of angina has been reviewed previously. The quantity and quality of this evidence is limited but there is consensus that BBs and CCBs are effective in the treatment of people with stable angina ^20–23.

The aim of this review was to determine whether BBs or CCBs offer advantages as first-line treatment for people with stable angina. The review includes evidence from nine trials of monotherapy with BBs versus monotherapy with CCBs.

Beta blockers

Beta blockers reduce myocardial oxygen consumption by competitive inhibition of beta-adrenoceptors, which lowers heart rate, blood pressure, and myocardial contractility. The bradycardia prolongs diastole, thereby increasing the period of maximal coronary blood flow. Relative contra-indications to beta-blockade include obstructive airways disease, acute heart failure, and impaired atrioventricular conduction. Side effects of BBs include fatigue, altered carbohydrate metabolism, peripheral vasoconstriction, sexual dysfunction, and bronchoconstriction. Some BBs (e.g. atenolol, metoprolol, bisoprolol) are relatively cardioselective with greater inhibition of the cardiac beta1 receptors than the beta2 (bronchial) receptors and therefore have less effect on airways resistance. Atenolol, bisoprolol, and nadolol have a relatively long duration of action and are given once daily. Other BBs with shorter half-lives may be given in slow-release formulations.

In the United Kingdom the most frequently prescribed BBs are atenolol, bisoprolol, and propranolol. The cost of a BB for four weeks is low (e.g. £0.99 for atenolol 50mg daily, £1.33 for bisoprolol 10mg daily)²⁴.

Calcium channel blockers

Calcium channel blockers inhibit movement of calcium through slow calcium channels of cell membranes in the myocardium, cardiac conduction tissues, and vascular smooth muscle. Calcium channel blockers dilate peripheral and coronary arteries, and to a varying degree depress myocardial contractility and intra-cardiac conduction. Calcium channel blockers include dihydropyridines (e.g. amlodipine), benzothiapines (e.g. diltiazem), and phenylalkylamines (e.g. verapamil). Dihydropyridines may cause reflex tachycardia, flushing, headache, and ankle swelling. Diltiazem and verapamil depress cardiac conduction and cause bradycardia, and should not be given to people with heart block or treated with a BB. Verapamil may cause constipation.

In the United Kingdom the most frequently prescribed CCBs are amlodipine, nifedipine, felodipine, diltiazem, and verapamil. The costs of CCBs for four weeks are higher than the costs of atenolol (e.g. amlodipine 10mg daily £1.43; diltiazem MR 60mg tds £2.93; verapamil 80mg tds £2.07)²⁴. Slow release formulations of CCBs are more expensive.

Generic beta blockers, calcium channel blockers included in evidence reviews

A large number of BBs and CCBs are available for clinical use in the UK and different BBs and CCBs have been used in different trials.

We looked at the prescription cost analysis table in the NHS Information Centre. We extracted the number of prescriptions in 2005 and 2008 for BB and CCB dispensed in the community. The GDG reviewed the lists and made a judgement about which drugs were currently used in stable angina and were known to be included in studies.

In this guideline we have considered evidence for the use of the following drugs:

BB: atenolol, propranolol, bisoprolol, metoprolol, nadolol,
CCB: amlodipine, diltiazem, felodipine, nifedipine, verapamil

In this review we have assumed that the clinical effects are consistent within a class of drug (e.g. BB or CCB), across a range of doses, and in all trial participants.

7.2. Beta blocker vs. calcium channel blocker

7.2.1. Clinical question

What is the comparative clinical/cost effectiveness of standard antianginal drugs (BBs/CCBs) for the management of angina?

7.2.2. Clinical evidence

The “Review Protocol” for this topic can be found in Appendix C, the “Search Strategies” in Appendix D, the “List of Included and Excluded Studies” in Appendix E1, the “Clinical Evidence Tables” in Appendix E2, and the “Forest Plots” in Appendix F.

Table 7.1BB vs. CCB for stable angina

Quality assessment							Summary of findings
Quality assessment							No of patients		Effect		Quality
No of studies	Design	Limitations	Inconsistency	Indirectness	Imprecision	Other considerations	BB	CCB	Relative (95% CI)	Absolute	Quality
Exercise duration (min) (metoprolol vs. diltiazem; propranolol vs. diltiazem; propranolol vs. nifedipine) (follow-up 6 weeks-6 months; better indicated by higher values)
van Dijk 1988²⁵; O’Hara 1987²⁶; Kawanishi 1992²⁷	randomised trials	serious ^(a)	no serious inconsistency	no serious indirectness	serious ^(b)	none	88	83	-	MD 0.05 higher (0.82 lower to 0.92 higher)	⊕⊕○○ LOW
Time to 1mm ST depression (sec) - metoprolol vs. nifedipine (follow-up 10 weeks; better indicated by higher values)
Savonitto 1996²⁸ (IMAGE)	randomised trials	serious ^(c)	no serious inconsistency	no serious indirectness	serious ^(b)	none	65	62	-	MD 12 higher (35.06 lower to 59.06 higher)	⊕⊕○○ LOW
Time to onset of angina (min) (metoprolol vs. dilitazem; propranolol vs. nifedipine) (follow-up 6 weeks-6 months; better indicated by lower values)
van Dijk 1988²⁵; Kawanishi 1992²⁷	randomised trials	serious ^(d)	no serious inconsistency	no serious indirectness	serious ^(b)	none	54	49	-	MD 0.63 higher (0.27 lower to 1.53 higher)	⊕⊕○○ LOW
Total mortality (atenolol vs. verapamil; metoprolol vs. verapamil; metoprolol vs. verapamil) (follow-up 2.7–3.4 years)
Pepine 2003²⁹ (INVEST); Rehnqvist 1996³⁰ (APSIS);	randomised trials	serious ^(e)	no serious inconsistency	no serious indirectness	no serious imprecision	none	915/11715 (7.8%)	898/11670 (7.7%)	RR 1.02 (0.93 to 1.11)	2 more per 1000 (from 5 fewer to 8 more)	⊕⊕⊕○ MODERATE
Cardiovascular death (atenolol vs. verapamil; atenolol vs. nifedipine; metoprolol vs. verapamil) (follow-up 2–3.4 years)
Pepine 2003²⁹ (INVEST); Dargie 1996³¹ (TIBET); Rehnqvist 1996³⁰ (APSIS)	randomised trials	serious ^(f)	no serious inconsistency	no serious indirectness	no serious imprecision	none	453/11941 (3.8%)	456/11902 (3.8%)	RR 0.99 (0.87 to 1.12)	0 fewer per 1000 (from 5 fewer to 4 more)	⊕⊕⊕○ MODERATE
Non fatal MI (atenolol vs. verapamil; atenolol vs. nifedipine; metoprolol vs. verapamil) (follow-up 2–3.4 years)
Pepine 2003²⁹ (INVEST); Dargie 1996³¹ (TIBET); Hjemdahl 2006³² (APSIS)	randomised trials	serious ^(g)	no serious inconsistency	no serious indirectness	no serious imprecision	none	184/11941 (1.5%)	185/11902 (1.6%)	RR 0.99 (0.81 to 1.22)	0 fewer per 1000 (from 3 fewer to 3 more)	⊕⊕⊕○ MODERATE
CV related hospitalisation – (atenolol vs. verapamil) (follow-up mean 2.7 years)
Pepine 2003²⁹ (INVEST)	randomised trials	no serious limitations ^(h)	no serious inconsistency	no serious indirectness	no serious imprecision	none	709/11309 (6.3%)	726/11267 (6.4%)	OR 0.97 (0.88 to 1.08)	2 fewer per 1000 (from 7 fewer to 5 more)	⊕⊕⊕⊕ HIGH
Non fatal CV events (combined) – (metoprolol vs. verapamil)(follow-up median 3.4 years)
Rehnqvist 1996³⁰ (APSIS)	randomised trials	serious ⁽ⁱ⁾	no serious inconsistency	no serious indirectness	serious ^(j)	none	106/406 (26.1%)	98/403 (24.3%)	RR 1.07 (0.85 to 1.36)	17 more per 1000 (from 36 fewer to 88 more)	⊕⊕○○ LOW
Angina episodes/week (atenolol vs. verapamil; metoprolol vs. diltiazem; propranolol vs. nifedipine; metoprolol vs. nifedipine) (follow-up 6 weeks-2.7 years; better indicated by lower values)
Pepine 2003²⁹ (INVEST); van Dijk 1988²⁵; Kawanishi 1992²⁷; Savonitto 1996²⁸ (IMAGE) ^(s)	randomised trials	serious ^(k)	no serious inconsistency	no serious indirectness	no serious imprecision	none	11424	11377	-	MD 0.11 higher (0.07 to 0.15 higher)	⊕⊕⊕○ MODERATE
Prevalence of angina – (atenolol vs. verapamil) (follow-up mean 2.7 years)
Pepine 2003²⁹ (INVEST)	randomised trials	no serious limitations ^(h)	no serious inconsistency	no serious indirectness	serious ^(j)	none	228/11309 (2%)	261/11267 (2.3%)	RR 0.87 (0.73 to 1.04)	3 fewer per 1000 (from 6 fewer to 1 more)	⊕⊕⊕○ MODERATE
Severity of angina assessed by investigator (moderate/markedly improved) – (nadolol vs. amlodipine) (follow-up 6 months)
Singh 1993³³	randomised trials	serious ^(l)	no serious inconsistency	no serious indirectness	no serious imprecision	none	21/39 (53.8%)	29/39 (74.4%)	RR 0.72 (0.51 to 1.02)	208 fewer per 1000 (from 364 fewer to 15 more)	⊕⊕⊕○ MODERATE
Severity of angina assessed by patients (moderate/severe) – (nadolol vs. amlodipine) (follow-up 6 months)
Singh 1993³³	randomised trials	serious ^(l)	no serious inconsistency	no serious indirectness	no serious imprecision	none	16/40 (40%)	12/40 (30%)	RR 1.33 (0.73 to 2.45)	99 more per 1000 (from 81 fewer to 435 more)	⊕⊕⊕○ MODERATE
Nitroglycerin use – (propranolol vs. nifedipine) (follow-up 6 months; better indicated by lower values)
Kawanishi 1992²⁷	randomised trials	serious ^(m)	no serious inconsistency	no serious indirectness	no serious imprecision	none	21	16	-	MD 0 higher (0.94 lower to 0.94 higher)	⊕⊕⊕○ MODERATE
Adverse effects (head ache) – (metoprolol vs. verapamil) (follow-up median 3.4 years)
Rehnqvist 1996³⁰ (APSIS)	randomised trials	serious ⁽ⁱ⁾	no serious inconsistency	no serious indirectness	no serious imprecision	none	3/406 (0.7%)	4/403 (1%)	RR 0.74 (0.17 to 3.31)	3 fewer per 1000 (from 8 fewer to 23 more)	⊕⊕⊕○ MODERATE
Adverse effects (GI events) – (metoprolol vs. verapamil) (follow-up median 3.4 years)
Rehnqvist 1996³⁰ (APSIS)	randomised trials	serious ⁽ⁱ⁾	no serious inconsistency	no serious indirectness	serious ⁽ⁿ⁾	none	10/406 (2.5%)	22/403 (5.5%)	OR 0.45 (0.22 to 0.94)	29 fewer per 1000 (from 3 fewer to 42 fewer)	⊕⊕○○ LOW
Adverse effects (dizziness) – (atenolol vs. verapamil) (follow-up mean 2.7 years)
Pepine 2003²⁹ (INVEST)	randomised trials	no serious limitations ^(h)	no serious inconsistency	no serious indirectness	no serious imprecision	none	151/11309 (1.3%)	154/11267 (1.4%)	RR 0.98 (0.78 to 1.22)	0 fewer per 1000 (from 3 fewer to 3 more)	⊕⊕⊕⊕ HIGH
Adverse effects (lightheadedness) – (atenolol vs. verapamil) (follow-up mean 2.7 years)
Pepine 2003²⁹ (INVEST)	randomised trials	no serious limitations ^(h)	no serious inconsistency	no serious indirectness	no serious imprecision	none	70/11309 (0.6%)	48/11267 (0.4%)	RR 1.45 (1.01 to 2.1)	2 more per 1000 (from 0 more to 5 more)	⊕⊕⊕⊕ HIGH
Adverse effects (overall) (atenolol vs. amlodipine; metoprolol vs. verapamil; nadolol vs. amlodipine) (follow-up 10 weeks-3.4years)
Pehrsson 2000³⁴; Rehnqvist 1996³⁰ (APSIS); Singh 1993³³^(q)	randomised trials	serious ^(o)	no serious inconsistency	no serious indirectness	no serious imprecision	none	139/562 (24.7%)	146/559 (26.1%)	RR 0.95 (0.79 to 1.14)	13 fewer per 1000 (from 55 fewer to 37 more)	⊕⊕⊕○ MODERATE
Adverse effects (constipation) – (atenolol vs. verapamil) (follow-up mean 2.7 years)
Pepine 2003²⁹ (INVEST)	randomised trials	no serious limitations ^(h)	no serious inconsistency	no serious indirectness	no serious imprecision	none	15/11309 (0.1%)	195/11267 (1.7%)	RR 0.08 (0.05 to 0.13)	16 fewer per 1000 (from 15 fewer to 16 fewer)	⊕⊕⊕⊕ HIGH
Withdrawals due to adverse effects – (atenolol vs. nifedipine) (follow-up mean 2 years)
Dargie 1996³¹ (TIBET) ^(r)	randomised trials	serious ^(p)	no serious inconsistency	no serious indirectness	serious ⁽ⁿ⁾	none	60/226 (26.5%)	93/232 (40.1%)	RR 0.66 (0.51 to 0.87)	136 fewer per 1000 (from 52 fewer to 196 fewer)	⊕⊕○○ LOW
Combined outcome (death, non fatal MI, non fatal stroke) (diabetes) - atenolol vs. verapamil (follow-up mean 2.7 years)
Pepine 2003²⁹ (INVEST)	randomised trials	no serious limitations ^(h)	no serious inconsistency	no serious indirectness	no serious imprecision	none	450/3231 (13.9%)	463/3169 (14.6%)	RR 0.95 (0.85 to 1.07)	7 fewer per 1000 (from 22 fewer to 10 more)	⊕⊕⊕⊕ HIGH
Combined outcomes (death, non fatal MI, non fatal stroke) (females) - atenolol vs. verapamil (follow-up mean 2.7 years)
Pepine 2003²⁹ (INVEST)	randomised trials	no serious limitations ^(h)	no serious inconsistency	no serious indirectness	no serious imprecision	none	540/5920 (9.1%)	524/5850 (9%)	RR 1.02 (0.91 to 1.14)	2 more per 1000 (from 8 fewer to 13 more)	⊕⊕⊕⊕ HIGH
Combined (death, non fatal MI, non fatal stroke) - subgroup age >70 - atenolol vs. verapamil (follow-up mean 2.7 years)
Pepine 2003²⁹ (INVEST)	randomised trials	no serious limitations ^(h)	no serious inconsistency	no serious indirectness	no serious imprecision	none	664/3829 (17.3%)	596/3694 (16.1%)	RR 1.07 (0.97 to 1.19)	11 more per 1000 (from 5 fewer to 31 more)	⊕⊕⊕⊕ HIGH
Quality of life (sleep disturbance) – (metoprolol vs. verapamil) (follow-up median 3.4 years; better indicated by lower values)
Rehnqvist 1996³⁰ (APSIS)	randomised trials	serious ⁽ⁱ⁾	no serious inconsistency	no serious indirectness	serious ^(b)	none	270	275	-	MD 0.4 lower (1.3 lower to 0.5 higher)	⊕⊕○○ LOW
Quality of life (overall life satisfaction) –(metoprolol vs. verapamil) (follow-up median 3.4 years; better indicated by lower values)
Rehnqvist 1996³⁰ (APSIS)	randomised trials	serious ⁽ⁱ⁾	no serious inconsistency	no serious indirectness	serious ^(b)	none	268	275	-	MD 0.7 lower (5.07 lower to 3.67 higher)	⊕⊕○○ LOW
Quality of life (psychosomatic symptoms) – (metoprolol vs. verapamil) (follow-up median 3.4 years; better indicated by lower values)
Rehnqvist 1996³⁰ (APSIS)	randomised trials	serious ⁽ⁱ⁾	no serious inconsistency	no serious indirectness	serious ^(b)	none	275	282	-	MD 1.3 lower (3.89 lower to 1.29 higher)	⊕⊕○○ LOW

a: van Dijk 1988²⁵; O’Hara 1987²⁶; Kawanishi 1992²⁷: All 3 studies randomised. Allocation concealment not reported in all 3 studies. All 3 studies double blind. ITT not reported in all 3 studies.
b: 95% CI includes no effect and the upper and lower CI crosses the MID.
c: Randomised. Double blind. Allocation concealment not reported. Baseline comparison made. Drop out <20% (11%). Intention to treat analysis not reported.
d: van Dijk 1988²⁵; Kawanishi 1992²⁷: Both studies randomised. Allocation concealment not reported in both studies. ITT not reported in both studies. Both studies double blind.
e: Pepine 2003²⁹; Hjemdahl 2006³² (APSIS); Rehnqvist 1996³⁰ (APSIS); All 3 randomised. Allocation concealment not reported in all 3 studies and ITT used in all 3 the studies. All 3 studies double blind.
f: Pepine 2003²⁹; Rehnqvist 1996³⁰ (APSIS); Dargie 1996³¹ (TIBET): All 3 studies randomised. Allocation concealment reported in 1 of the 3 studies. ITT reported in all 3 studies. All 3 studies double blind.
g: Dargie 1996 (TIBET); Pepine 2003²⁹; Hjemdahl 2006³² (APSIS): All 3 studies randomised. Allocation concealment reported in 1 of 3 studies. ITT reported in all 3 studies. All 3 studies double blind.
h: Randomised. Allocation concealment reported. Double blind. Power calculation reported. Drop-out rate <20% (2.5%). Baseline comparisons made. Intention to treat analysis used.
i: Double blind. Randomised. Allocation concealment not reported. Baseline comparisons made. Power calculation reported. Drop out <20%. Intention to treat analysis reported.
j: 95% CI around the pooled estimate of effect includes both: 1) no effect and 2) appreciable benefit or appreciable harm.
k: Pepine 2003²⁹; van Dijk 1988²⁵; Kawanishi 1992²⁷; Savonitto 1996²⁸: All 4 studies randomised. Allocation concealment not reported in 3 of the 4 studies. ITT not reported in 3 of the 4 studies. All 4 studies double blind.
l: Double blind. Randomised. Allocation concealment not reported. Baseline comparisons made. Drop out >20% 23% [(19/80) drop out; 20% (8/40) in the amlodipine group and 27% (11/40) in the nadolol group]. Intention to treat analysis not reported.
m: Randomised. Double blind. Allocation concealment not reported. Baseline comparisons made. Drop out < 20% (2.8% in nifedipine group and 2.6% in propronolol group). Intention to treat analysis not reported.
n: 95% CI around the pooled estimate of effect includes appreciable benefit or appreciable harm.
o: Pehrsson 2000³⁴; Rehnqvist 1996³⁰ (APSIS); Singh 1993³³: Randomised all 3 studies. Allocation concealment not reported in all 3 of the studies. ITT not reported in 2 of the 3 studies. All 3 studies double blind.
p: Randomised. Double blind. Allocation concealment not reported. Baseline comparisons made. Drop-out >20% [60(27%) for atenolol, 93 (40%) for nifedipine, 64 (29%) for their combination]. Intention to treat analysis reported.
q: Most commonly reported side effects with nadolol were bradycardia, dizziness, headache, nausea, dyspnoea, palpitations, and fatigue. Most frequently reported side effects with amlodipine were headache, oedema, palpitations, hypoesthesia, and flushing.
r: Not reported what were the side effects.
s: The comparisons here are metoprolol +placebo vs. nifedipine +placebo.

Additional data

Vliegen 1991³⁵

Population: n=56 (n=26 metoprolol, n=30 diltiazem).

People with stable effort induced angina pectoris for at least 3 months. The mean age in the diltiazem group was 58±9 yrs and in the metoprolol group was 64±9 yrs (p<0.05);

Intervention: metoprolol 100 mg b.i.d.

The treatment was preceded by a 2 week run-in period. If the patients were already taking antianginal medication (other than short-acting nitrates) this was gradually discontinued. In the second week of the run-in period, only short-acting nitrates were used by all patients. If the patients were not taking antianginal medication, the single blind run-in period was 1 week.

Comparison: diltiazem 120 mg b.i.d.

Follow-up: Follow-up 8 weeks, 20 weeks and 32 weeks.

Results:

Exercise test (32 weeks): during treatment, mean changes in duration of exercise, time to angina pectoris, time to 1 mm ST segment depression, maximal ST segment depression were not significantly different between the patients on diltiazem and those on metoprolol. However at 20 weeks, exercise duration was longer in patients on diltiazem than in patients on metoprolol.
Frequency of angina (8 weeks): the mean frequency of anginal attacks/week decreased in diltiazem group from 5.9 at baseline to 3.5 during treatment (p<0.05) and in the metoprolol group from 7.4 at baseline to 4.7 during treatment (p<0.01). No differences were observed between the two treatment groups.

Side effects: no significant differences were found in incidence and severity of side effects between the 2 groups.

Drug dosages in each study

Dargie 1996³¹ (TIBET) - atenolol 50 mg twice daily, nifedipine (slow release) 20–40 mg twice daily
Pepine 2003²⁹ (INVEST) - Group 1: atenolol 50 mg twice daily + hydrochlorothiazide 25 mg twice daily + trandolapril 2mg/d; Group 2: verapamil sustained release, 180 mg twice daily + hydrochlorothiazide, 25 mg/d + trandolapril, 2 mg twice daily
Pehrsson 2000³⁴ - amlodipine 10 mg once daily, atenolol 100 mg once daily.
van Dijk 1988²⁵ - diltiazem 240 mg (60 mg four times daily), metoprolol 200 mg (100 mg twice daily)
Savonitto 1996²⁸ (IMAGE)- metoprolol (controlled release, 200 mg once daily), nifedipine (retard, 20 mg tablets twice daily)
Rehnqvist 1996³⁰ (APSIS), Hjemdahl 2006³² (APSIS)- metoprolol (Seloken ZOC 200 mg once daily), verapamil (Isoptin Retard 240 twice daily)
Singh 1993³³ - amlodipine 2.5–10 mg once daily, nadolol 40–160 mg once daily
O’ Hara 1987²⁶ - diltiazem 360 mg once daily, propranolol 240 mg once daily
Kawanishi 1992²⁷ - nifedipine 10 mg four times daily vs. propranolol 20mg four times daily (not specified if it is long- or short-acting nifedipine)

7.2.3. Economic evidence

One study³⁶ included the relevant comparison. This is summarised in the economic evidence profile below. In this cost study from the UK, angina-related healthcare resource use over one year for angina patients who received one of the included drugs was obtained from a UK longitudinal database. Three subgroup analyses were conducted where resource use was monitored for 12 months after: a) patients commenced angina treatment for the first time b) patients were switched to a different angina treatment c) patients had received the same angina treatment for at least one year. Unit costs were obtained from published literature and NHS databases and were attached to resources. No clinical outcome was evaluated. The base case results reported are for patients without any of the following comorbidities: ischaemic heart disease (excluding angina), hypertension, congestive cardiac failure, hypercholesterolaemia and cerebrovascular disease. We report the results for patients with comorbidity as a part of sensitivity analysis. See also Economic Evidence Tables in Appendix G.

Table 7.2BB vs. CCB – Economic study characteristics

Study	Limitations	Applicability	Other Comments
Borghi 2000³⁶	Potentially serious limitations ^(a)	Partial applicability ^(b)	Atenolol and diltiazem were respectively the BB and CCB evaluated. Resource use data were obtained from a database.

a: Based on a cross-sectional study; only one drug from each group was evaluated, the range of GP visits frequency used in the sensitivity analysis is not reported.
b: Not a full economic evaluation: only costs, not health effects.

Table 7.3BB vs. CCB – Economic summary of findings

Study	Incremental cost per patient (£)	Incremental effects	ICER	Uncertainty
Patients in their first year of antianginal treatment
Borghi 2000³⁶	Saves 358 ^(a)	NR	NA	Subgroup analysis in patients with comorbidities: BB has an incremental cost of £580 per patient. The main cost drivers were the acquisition cost of the anti-anginal drugs, hospitalisation and GP visits and these were varied in a sensitivity analysis. The overall results do not change when: - frequency of GP visits is varied - incidence of hospitalisation is varied (from 0 to double) - the cost of generic drugs is used.
Patients in the year following a change in previous medication
Borghi 2000³⁶	97 ^(a)	NR	NA	Subgroup analysis in patients with comorbidities: BB has an incremental cost of £232 per patient.. The main cost drivers were the acquisition cost of the anti-anginal drugs, hospitalisation and GP visits and these were varied in a sensitivity analysis. The overall results do not change when: - frequency of GP visits is varied - incidence of hospitalisation is varied (from 0 to double) - the cost of generic drugs is used.
Patients who had received the same treatment during the previous year
Borghi 2000³⁶	Saves 16 ^(a)	NR	NA	Subgroup analysis in patients with comorbidities: BB saves £221 per patient. The main cost drivers were the acquisition cost of the anti-anginal drugs, hospitalisation and GP visits and these were varied in a sensitivity analysis. The overall results do not change when: - frequency of GP visits is varied - incidence of hospitalisation is varied (from 0 to double) - the cost of generic drugs is used.

a: 1997/1998 GBP. Costs included were cost of anti-anginal drugs, additional medication, GP-initiated tests, GP and practice nurse visits, outpatient visits, elective and emergency admissions. Resource costs were obtained NHS databases and UK cost studies.

7.2.4. Evidence statements

Clinical

BB vs. CCB

Clinical efficacy:

Pepine 2003²⁹ (INVEST), Van Dijk 1988²⁵, Kawanishi 1992²⁷, Savonitto 1996²⁸ (IMAGE): Evidence from 4 RCT’s shows that there were significantly fewer anginal episodes/week [MD 0.11 (0.07 to 0.15)] with CCB (verapamil, diltiazem, nifedipine) compared with BB (atenolol, metoprolol, propranolol) (follow-up 6 weeks–2.7 years).

Van Dijk 1988²⁵, O’Hara 1987²⁶, Kawanishi 1992²⁷: Evidence from 3 RCTs shows that there was no significant difference between BB (metoprolol, propranolol) and CCB (diltiazem, nifedipine) for exercise duration (min) [MD 0.05 (−0.82 to 0.92)] (follow-up 6 weeks–6 months).

Savonitto 1996²⁸ (IMAGE): Evidence from one RCT shows that there was no significant difference between BB (metoprolol) and CCB (nifedipine) for time to 1mm ST segment depression [MD 12 (−35.06 to 59.06)] (follow-up 10 weeks).

Van Dijk 1988²⁵, Kawanishi 1992²⁷: Evidence from 2 RCTs shows that there was no significant difference between BB (metoprolol, propranolol) and CCB (diltiazem, nifedipine) for time to onset of angina (min) [MD 0.63 (−0.27 to 1.53)] (follow-up 6 weeks–6 months).

Pepine 2003²⁹ (INVEST), Rehnqvist 1996³⁰ (APSIS): Evidence from 2 RCTs shows that there was no significant difference between BB (atenolol, metoprolol) and CCB (verapamil) for total mortality [RR 1.02 (0.93 to 1.11)]. (follow-up 2.7–3.4 years)

Pepine 2003²⁹ (INVEST), Dargie 1996³¹ (TIBET), Rehnqvist 1996³⁰ (APSIS): Evidence from 3 RCTs shows that there was no significant difference between BB (atenolol, metoprolol) and CCB (verapamil, nifedipine) for cardiovascular death [RR 0.99 (0.87 to 1.12)] (follow-up 2–3.4 years).

Pepine 2003²⁹ (INVEST), Dargie 1996³¹ (TIBET), Hjemdahl 2006³² (APSIS): Evidence from 3 RCTs shows that there was no significant difference between BB (atenolol, metoprolol) and CCB (verapamil, nifedipine) for non fatal MI [RR 0.99 (0.81 to 1.22)] (follow-up 2–3.4 years).

Pepine 2003²⁹ (INVEST): Evidence from one RCT shows that there was no significant difference between BB (atenolol) and CCB (verapamil) for cardiovascular related hospitalisation [RR 0.97 (0.88 to 1.08)] (follow-up mean 2.7 years).

Rehnqvist 1996³⁰ (APSIS): Evidence from one RCT shows that there was no significant difference between BB (metoprolol) and CCB (verapamil) for non fatal CV events (acute MI, incapacitating or unstable angina, cerebrovascular events or peripheral vascular events). [RR 1.07 (0.85 to 1.36)] (Follow-up median 3.4 years).

Pepine 2003²⁹ (INVEST): Evidence from one RCT shows that there was no significant difference between BB (atenolol) and CCB (verapamil) for prevalence of angina [RR 0.87 (0.73 to 1.04)] (follow-up mean 2.7 years).

Singh 1993³³: Evidence from one RCT shows that there was no significant difference between BB (nadolol) and CCB (amlodpine) for severity of angina (assessed by investigators as moderate/markedly improved) [RR 0.72 (0.51 to 1.02)] (follow-up 6 months).

Singh 1993³³: Evidence from one RCT shows that there was no significant difference between BB (nadolol) and CCB (amlodipine) for severity of angina (assessed by patients as moderate/severe) [RR 1.33 (0.73 to 2.45)] (follow-up 6 months).

Kawanishi 1992²⁷: Evidence from one RCT shows that there was no significant difference between BB (propranolol) and CCB (nifedipine) for use of nitroglycerin tablets/week [MD 0 (−0.94 to 0.94)] (follow-up 6 months).

Rehnqvist 1996³⁰ (APSIS): Evidence from one RCT shows that there was no significant difference between BB (metoprolol) and CCB (verapamil) for quality of life psychosomatic symptoms [MD −1.3 (−3.89 to 1.29)], overall life satisfaction [MD −0.7 (−5.07 to 3.67)], and sleep disturbance [MD −0.4 (−1.3 to 0.5)] [ (follow-up median 3.4 years).

Pepine 2003²⁹ (INVEST): Evidence from one RCT shows that there was no significant difference between BB and CCB for combined outcomes (death, non fatal MI, non fatal stroke) in sub group analyses conducted for age> 70 years [RR 1.07 (0.97 to 1.19)], female gender [RR 1.02 (0.91 to 1.14)] and people with diabetes [RR 0.95 (0.85 to 1.07)] (follow-up mean 2.7 years).

Adverse effects:

Dargie 1996³¹ (TIBET): Evidence from one RCT shows that there were significantly more withdrawals due to adverse effects [RR 0.66 (0.51 to 0.87)] with CCB (nifedipine) compared to BB (atenolol) (follow-up mean 2 years).

Pepine 2003²⁹ (INVEST): Evidence from one RCT shows that there were significantly more adverse effects (constipation) [RR 0.08 (0.05 to 0.13)] with CCB (verapamil) compared to BB (atenolol) (follow-up mean 2.7 years).

Pepine 2003²⁹ (INVEST): Evidence from one RCT shows that there were significantly more adverse effects (light headedness) [RR 1.45 (1.01 to 2.1)] with BB (atenolol) compared to CCB (verapamil) (follow-up mean 2.7 years).

Rehnqvist 1996³⁰ (APSIS): Evidence from one RCT shows that there were significantly more adverse effects (GI events) [RR 0.45 (0.22 to 0.94)] with CCB (verapamil) compared to BB (metoprolol) (median 3.4 years).

Pehrsson 2000³⁴, Rehnqvist 1996³⁰ (APSIS), Singh 1993: Evidence from 3 RCTs shows that there was no significant difference between BB (atenolol, metoprolol, nadolol) and CCB (amlodipine, verapamil) for adverse effects (overall) [RR 0.95 (0.79 to 1.14)] (follow-up 10 weeks–3.4 years).

Pepine 2003²⁹ (INVEST): Evidence from one RCT shows that there was no significant difference between BB (atenolol) and CCB (verapamil) in adverse effects (dizziness) [RR 0.98 (0.78 to 1.22)] (follow-up mean 2.7 years).

Rehnqvist 1996³⁰ (APSIS): Evidence from one RCT shows that there was no significant difference between BB (metoprolol) and CCB (verapamil) for adverse effects (head ache) [RR 0.74 (0.17 to 3.31)] (follow-up median 3.4 years).

Economic

Patients with and without co-morbidities were analysed separately. In patients without comorbidities BB generate fewer costs during the first year of treatment. BB and CCB have similar costs after the first year. In patients with comorbidities BB generate more costs also during the first year. This evidence has potentially serious limitations and partial applicability.

7.2.5. Recommendations and link to evidence


*Recommendation*	Offer either a beta blocker or a calcium channel blocker as first-line treatment for stable angina. Decide which drug to use based on comorbidities, contraindications and the person’s preference. If the person cannot tolerate the beta blocker or calcium channel blocker, consider switching to the other option (calcium channel blocker or beta blocker). Do not routinely offer anti-anginal drugs other than beta blockers or calcium channel blockers as first-line treatment for stable angina.

Relative values of different outcomes	The outcomes of most interest were long-term mortality (total and cardiovascular) and rates of major adverse cardiovascular events (myocardial infarction, stroke, myocardial revascularisation). Other outcomes included were measures of symptom severity (frequency of angina, exercise test outcomes).
Trade off between clinical benefits and harms	We found no evidence of a difference in total or cardiovascular mortality, or in risk of myocardial infarction or stroke, between people with stable angina treated with CCB or BB. In one large trial the effect of treatment with CCB and BB on a combined endpoint (death, non-fatal myocardial infarction, non-fatal stroke) was consistent across subgroups, including women, and people with diabetes or aged over 70 years. In one high quality trial the prevalence of angina two years after randomisation was similar amongst people treated with sustained release verapamil and amongst people treated with atenolol. On the other hand, evidence from four randomised controlled trials suggests that there are 0.11 fewer angina episodes per week amongst patients treated with CCB than amongst patients treated with BB. This difference equates to a single episode of angina every nine weeks and the GDG did not consider this to be of major clinical significance. In one trial there was no difference in quality of life assessed with the Cornell Medical Index between patients treated with CCB or with BB. There is no evidence of a consistent and clinically important difference in the rate of adverse events between patients treated with BB or CCB. In one large trial treatment with verapamil was associated with constipation but treatment with atenolol was associated with light-headedness. The GDG concluded that there is no evidence to discriminate between BB and CCB for the initial treatment of people with stable angina. Evidence to guide treatment if monotherapy with a BB or a CCB is not tolerated or does not control symptoms of angina is very limited. The GDG reached a consensus that if one class of anti-anginal drug is not tolerated or is ineffective a switch to the other class of anti-anginal drug can be considered.
Economic considerations	The cost of treatments with BB and CCB and their consequences is similar after the first year. The presence of comorbidities might influence the level of resource use (e.g. admissions) during the first year.
Quality of evidence	Randomised trials of BBs and CCBs in people with stable angina have mainly studied older drugs within each drug class (e.g. propranolol, atenolol, metoprolol, nifedipine, diltiazem, verapamil). The trials selectively recruited patients who were suitable for treatment with either a BB or CCB. Information about the long term effects of BBs and CCBs in the treatment of people with stable angina is very limited. Most trials were not designed to study the effects of treatment on mortality or other major cardiovascular outcomes, are limited by small study size, and only report short to medium term follow-up. One large trial was designed to detect a difference in the composite rate of death, non-fatal myocardial infarction, and non-fatal stroke at two years.²⁹ The economic evidence has potentially serious limitations (it was based on a cross-sectional study and only one drug from each group was evaluated) and partial applicability (only costs, not health effects were measured).
Other considerations	The GDG recognised the historical consensus that monotherapy with BB or CCB is effective for the prevention of attacks of angina. The GDG was also aware that monotherapy with organic nitrates is limited by the development of tolerance, and that evidence to support monotherapy with other antianginal drugs (nicorandil, ivabradine, ranolazine) is very limited. The GDG concluded that anti-anginal drugs other than BBs or CCBs should not be used as first line treatments for stable angina. Previous guidelines^20–22 have suggested that BBs should be the first-line treatment for stable angina because of evidence that beta-blockade reduces mortality after acute myocardial infarction³⁷ and in people with chronic heart failure³⁸^,³⁹. It has also been suggested that short-acting dihydropyridines may have deleterious effects in people with coronary artery disease⁴⁰. We found no evidence to differentiate between the use of BB versus the use of CCB for first-line treatment of stable angina. The GDG were also aware of a consensus that BBs and CCBs have a class effect on symptoms of angina, but that the potential for a particular drug to cause adverse effects may be influenced by its pharmacological profile (for example cardioselectivity for BBs and effects on intra-cardiac conduction for CCBs). The GDG considered that the available evidence did not support a recommendation to use a specific BB or CCB. Nevertheless, clinicians should be aware that evidence for anti-anginal efficacy is mainly confined to the use of a small number of older agents (e.g. propranolol, atenolol, metoprolol, nifedipine, diltiazem, verapamil), and clinicians should consider comorbidity, contra-indications and patient preference when selecting a first-line anti-anginal agent. The difference in cost between BBs and CCBs is small and was not considered significant by the GDG. The choice of initial treatment should therefore be determined by co-morbidity, contraindications and patient preference.

Apart from any fair dealing for the purposes of research or private study, criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, no part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

The rights of the National Clinical Guidelines Centre to be identified as Author of this work have been asserted by them in accordance with the Copyright, Designs and Patents Act, 1988.

Bookshelf ID: NBK83601

Contents

< Prev Next >

PubReader
Print View
Cite this Page
National Clinical Guidelines Centre (UK). Stable Angina: Methods, Evidence & Guidance [Internet]. London: Royal College of Physicians (UK); 2011 Jul. (NICE Clinical Guidelines, No. 126.) 7, Beta blockers vs. calcium channel blockers.
PDF version of this title (3.7M)

Other titles in this collection

National Institute for Health and Clinical Excellence: Guidance

Related NICE guidance and evidence

Recent Activity

Clear Turn Off Turn On

Beta blockers vs. calcium channel blockers - Stable Angina
Beta blockers vs. calcium channel blockers - Stable Angina

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

Bookshelf