Beta-blockers for a weaker heart pump
Metoprolol succinate extended-release • carvedilol • bisoprolol
Detailed source records support the one-page handout. Different trials, populations, follow-up periods and stopping definitions cannot be compared as a drug ranking.
Scope and trial interpretation
Use one established beta-blocker for adult HFrEF. This page covers metoprolol succinate ER, immediate-release carvedilol and single-ingredient bisoprolol. U.S. bisoprolol labeling covers hypertension; HFrEF is guideline-supported off-label use. [1] [2] [3] [8] [7]
All three mortality trials stopped early for benefit and predate ARNI/SGLT 2 therapy. MERIT-HF had a placebo run-in, not active tolerator selection. COPERNICUS excluded marked fluid retention and active IV inotrope/vasodilator treatment. The different populations and follow-up prevent cross-trial ranking. [1] [2] [3] [8] [7]
Clinician-directed starting/target doses: metoprolol succinate 12.5–25/200 mg daily; carvedilol 3.125 mg twice daily/25–50 mg twice daily (severe-HF trial target 25 mg twice daily); bisoprolol 1.25/10 mg daily. Titrate stable compensated HF according to pulse, BP, symptoms and volume; no universal patient hold threshold. [1] [2] [3] [8] [7]
Stopping definitions and unresolved data
MERIT-HF JAMA reports any-cause withdrawal 279/1990 vs 310/2001, AE withdrawal 196/1990 vs 234/2001. Label 10.3%/12.2% adverse-reaction discontinuation is not numerically reconciled; use one source consistently. Patients could have multiple reasons. [4] [5] [2] [6] [3] [7]
COPERNICUS 14.8%/18.5% are Kaplan–Meier 1 year estimates excluding death, not observed 171/1156 vs 210/1133 counts. Overall AE-only numerator was not verified; label says AE discontinuation similar and dizziness-specific stopping 1.3%/0.6%. [4] [5] [2] [6] [3] [7]
CIBIS-II 194/1327 vs 192/1320 are varied-reason withdrawal, not AE-only. Overall HFrEF AE-only stopping was not verified; short hypertension label trials cannot substitute. Bradycardia hospitalization 14/1327 vs 2/1320 is a different endpoint. [4] [5] [2] [6] [3] [7]
Safety and missed doses
No abrupt cessation; review asthma, AV block, diabetes, pregnancy/lactation and rate-slowing drugs including eye drops, digoxin, amiodarone, verapamil/diltiazem. Carvedilol is contraindicated in asthma. Beta-1 selectivity does not eliminate bronchospasm. Hypoglycemia may lack tachycardia, especially with fasting/vomiting. [5] [6] [7]
Metoprolol ER: no crushing/chewing; scored tablets may be divided if instructed. Toprol-XL §17 missed dose: next scheduled dose only. Carvedilol patient information: take when remembered unless next dose due, never double; with food. No equivalent missed-dose instruction in reviewed U.S. bisoprolol label: follow product/pharmacist advice rather than a pooled rule. [5] [6] [7]
Observational persistence
Norwegian cohort 54899, first coded HF 2014–2020, age 18 to under 80, survived 30 days. Initiator KM persistence 72%/58%/38% at 1/2/5 years; 30 day grace, deathcensored, within-class switches. A 90 day grace raised estimates. No reasons for gaps; claims do not prove ingestion. Not restricted to HFrEF or these three drugs. Exact persistence-analysis N unavailable; do not use adherence sample 33106. [9]
Verified full-fill costs
GoodRx table examples: succinateER 25 mg×90 $15.75; carvedilolIR 12.5 mg×180 $20.85; bisoprolol 5 mg alone×30 $9.00. At 1/2/1 tablets daily:90/90/30 days. Pickup estimates, no paid Companion or mail shipping; not guaranteed local quotes or starting doses. [10] [11] [12] [13] [14] [15] [16] [17]
Verified Blueberry calculator 09/05/26: same strengths/quantities $18.45/$18.70/$12.07 before$3.99 shipping; individually shipped$22.44/$22.69/$16.06. Professional fee$15 for 90 day/$10 for 30 day, plus$1.15 supply and acquisition with markup. Manufacturer/NDC matters. No membership or lowest-price claim; combined shipment might share shipping. [10] [11] [12] [13] [14] [15] [16] [17]
KFF is a general employer-tier benchmark. Medicare requires actual plan price/stage; do not apply 25% to cash prices or reuse Entresto/Jardiance negotiated-price examples. Discount-card purchases bypassing PartD do not count toward deductible/cap. [10] [11] [12] [13] [14] [15] [16] [17]
Numerical records
MERIT-HF: Death from any cause
Stable NYHAII–IV;EF≤40%; Mean 1 year. Observed events/randomized N; placebo plus usual care; not annualized rates.
Metoprolol succinate ER: 145/1,990 (7.3%); Placebo: 217/2,001 (10.8%). [1]
COPERNICUS: Death from any cause
Euvolemic severe HF;EF<25%; Mean 10.4 months. Observed events/randomized N; placebo plus usual care; distinct from Kaplan–Meier mortality.
Carvedilol: 130/1,156 (11.2%); Placebo: 190/1,133 (16.8%). [2]
CIBIS-II: Death from any cause
Stable NYHAIII–IV;EF≤35%; Mean 1.3 years. Observed events/randomized N; placebo plus standard care.
Bisoprolol: 156/1,327 (11.8%); Placebo: 228/1,320 (17.3%). [3]
MERIT-HF: Permanent withdrawal, any cause
Randomized population; Mean 1 year. Any cause, not AE-only or leaving follow-up.
Metoprolol succinate ER: 279/1,990 (14.0%); Placebo: 310/2,001 (15.5%). [4]
MERIT-HF: Permanent withdrawal due to any adverse event
Randomized population; Mean 1 year. Exact JAMA counts; current label 10.3%/12.2% not reconciled.
Metoprolol succinate ER: 196/1,990 (9.8%); Placebo: 234/2,001 (11.7%). [4]
CIBIS-II: Permanent treatment withdrawal
Randomized population; Mean 1.3 years. Includes clinical need for beta-blocker, intolerance, other reasons; not AE-only.
Bisoprolol: 194/1,327 (14.6%); Placebo: 192/1,320 (14.5%). [3]
COPERNICUS: Permanent stopping for adverse effects or other nondeath reasons
Randomized baseline N 1156/1133; Kaplan–Meier at 1 year; mean trial follow-up 10.4 months. Estimated time-to-event percentage; exact observed counts not supplied. Do not infer 171/210.
Carvedilol: 14.8% estimated at 1 year; Placebo: 18.5% estimated at 1 year. [2]
MERIT-HF: Dizziness/vertigo
HF trial participants in label table; Mean 1 year. Rounded reported percentages regardless of causality; exact event counts not supplied; not valid cross-trial safety ranking.
Metoprolol succinate ER: 1.8%; Placebo: 1%. [5]
MERIT-HF: Bradycardia
HF trial participants in label table; Mean 1 year. Rounded reported percentages regardless of causality; exact event counts not supplied; not valid cross-trial safety ranking.
Metoprolol succinate ER: 1.5%; Placebo: 0.4%. [5]
COPERNICUS: Dizziness
HF trial participants in label table; Median medication exposure 10.4 months. Rounded reported percentages regardless of causality; exact event counts not supplied; not valid cross-trial safety ranking.
Carvedilol: 24%; Placebo: 17%. [6]
COPERNICUS: Bradycardia
HF trial participants in label table; Median medication exposure 10.4 months. Rounded reported percentages regardless of causality; exact event counts not supplied; not valid cross-trial safety ranking.
Carvedilol: 10%; Placebo: 3%. [6]
COPERNICUS: Hypotension
HF trial participants in label table; Median medication exposure 10.4 months. Rounded reported percentages regardless of causality; exact event counts not supplied; not valid cross-trial safety ranking.
Carvedilol: 14%; Placebo: 8%. [6]
COPERNICUS: Diarrhea
HF trial participants in label table; Median medication exposure 10.4 months. Rounded reported percentages regardless of causality; exact event counts not supplied; not valid cross-trial safety ranking.
Carvedilol: 5%; Placebo: 3%. [6]
CIBIS-II: Hospital admission for bradycardia
Randomized participants; Mean 1.3 years. Hospitalization, not all bradycardia reports or withdrawal.
Bisoprolol: 14/1,327 (1.1%); Placebo: 2/1,320 (0.2%). [3]
Norwegian HF registry 2014–2020: Continued beta-blocker refill persistence
Initiators within cohort 54899; exact analysis baseline N not given in main text; all EF phenotypes and beta-blockers; Kaplan–Meier at 2 years. Gaps≤30 days and within-class switches allowed; deathcensored. Refills not ingestion, AE attribution or irreversible stopping.
Beta-blocker initiators: 58% persistent at 2 years. [9]