Frans H. Rutten, Nicolaas P. A. Zuithoff, EelkoHak, Diederick E. Grobbee, Arno W. Hoes Arch Intern Med. 2010;170(10):880-887. Beta-blockers may reduce.

Frans H. Rutten, Nicolaas P. A. Zuithoff, EelkoHak, Diederick E. Grobbee, Arno W. Hoes

Arch Intern Med. 2010;170(10):880-887.

Beta-blockers may reduce mortality and risk of exacerbations in patients with chronic

obstructive pulmonary disease

Introduction: Prevalence and Definition

4th leading cause of death in the US, also a major cause of diabilityEstimated 12 million in US diagnosed, another estimated 12 million in US remain undiagnosed (http://www.nhlbi.nih.gov/health/public/lung/copd/index.htm)

Key components of the COPD definition by The GOLD (Global Initiative for Chronic Obstructive Lung Disease): (http://www.goldcopd.com/)

• preventable and treatable disease significant extrapulmonary effects • pulmonary component is characterized by airflow limitation that is

not fully reversible– airflow limitation is usually progressive and associated with an abnormal

inflammatory response of the lungs to noxious particles or gases

Introduction: Diagnosis

Suspect in all patients with chronic cough, chronic sputum production, dyspnea (at rest or with exertion), and/or history of inhalational exposure

Confirm with spirometry:GOLD: FEV1/FVC <0.70Hopkins: FEV1/FVC actually differs 5 or more percent from

predicted

Note: make sure no alternative diagnosis such as bronchiectasis, vocal cord paralysis, or tracheal stenosis which can mimic PFTs of COPD

Introduction: Therapy

http://www.spirometrie.info/img/gold_therapy.jpg

Introduction: Question: Does long-term beta blocker use improve survival and reduce the risk

of exacerbations in patients with COPD?

• Beta blockers improve survival in patients with heart failure, IHD

• Meta-analyses show that cardioselective beta-blockers are well-tolerated in COPD; no significant effect of FEV1, beta-agonist response, inhaler use, or respiratory symptoms

Salpeter SS, Ormiston T, Salpeter E, Poole P, Cates C. Cardioselective beta-blockers for chronic obstructive pulmonary disease. Cochrane Database Syst Re. 2002;(2):CD003566.

• Prior study showing that beta-blockers had a non-significant tendency to reduce all-cause mortality in patients with HTN and COPD

Au DH, Bryson CL, Fan VS, et al. Beta-blockers as single-agent therapy hypertension and the risk of mortality among patients with chronic obstructive pulmonary disease. Am J Med. 2004;117(12):925-931.

Methods: Study Design•Goal: Does long-term B-blocker use improve survival and reduce risk of COPD exac in COPD pts including those without CV disease.

• Observational Cohort of 2230 patients from the GP network database in Utrecht, Netherlands.

• 23 practices – 35 GPs from 1995 to 2005.

• Inclusion criteria: >= 45yrs, incident or prevalent diagnosis of COPD.

•Exclusion criteria: Patients who moved or lost to f/u, database doesn’t have nursing home residents.

• Outcomes: All cause mortality, 1st COPD exac during study period.

Methods: Definitions• COPD : by ICD 9 codes for chronic bronchitis,

COPD/emphysema, sx of dyspnea/cough/sputum for at least 3 mon/yr for 2 consecutive yrs and rhonchii on exam (70% of COPD cases conform to GOLD criteria).

• COPD exac: pulsed-dose prescription of steroids during 7-10 days and/or hospitalization for exac.

• Overt CV disease: Angina, MI, CABG, PCI, Afib, CHF, PAD, CVA, DM (Htn not included)

Methods: Data Analysis

• Cox proportional hazards regression used to calculate crude and adjusted hazard ratios for risk of all cause death and COPD exac with use of all beta blockers and then with cardioselective and non-cardioselective beta blockers.

• Missing smoking data: performed imputation and sensitivity analyses to compare to imputation data – were similar

• To adjust for confounding medication and improving power, they used Propensity score and Subgroup analysis

Patient Characteristics

-Small % of non-HTN pts in no β-blocker group

-More pts with CV comorbidities in the β-blocker group

-Could pts with CV dz not on β-blocker be undertreated? Could they represent more severe COPD?

Patient Characteristics

-Significantly more pts NOT on β-blocker ARE on inhaled β2 agonist

-More β-blocker pts treated with anticholinergics

-Are pts w/ COPD being treated for COPD appropriately?

-Do pts on inhaled β2 agonists but not on β-blockers have more severe COPD?

Mortality and β-blocker use

All CI < 1 All CI cross 1;higher pointestimates

Mortality and subgroup analysis

• Subgroups included:– No overt cardiovascular disease

(defined as no angina, MI, ischemic heart dz, afib, CHF, CVA, PAD; note HTN is missing)

– Pts taking meds:• 2 or more pulmonary drugs • β2 agonists• inhaled anticholinergics

– Incident cases of COPD– Pts referred to pulmonologist

Subgroup analysis“Patients without (overt) cardiovascularcomorbidities in our study still had HTNas the reason for β-blocker use”

-Note very low mortality rate, 19.6%

-All other subgroups had mortality rangingfrom 29.8 – 40.5%

-Implies healthier patients, greaterfunctional reserve, fewer CV Comorbidities

-Incident cases of COPD may indicate less severe disease at onset

-75% of pts included were “incident cases”

-Could incident COPD mean less severe?

Subgroup analysis

-Surprisingly few patients on β2 agoniststreated with β-blocker

-Is use of 2 or more pulmonary drugs really a surrogate of severe COPD? (i.e. β-agonist and inhaled steroids)

-Again, are pts being treated optimally for their underlying COPD?

Survival in COPD patients according to β-blocker use

No significant mortality effectuntil >20 months after initiation

Change in survival rate at ~60 monthsparallel to no β-blocker

Subgroup analysis

-Referral to pulmonologist as surrogate for severity of COPD

-β-blocker shows no improvement in mortality in pts expected to have severe COPD

-No convincing evidence of harm either

-Note that all of the CI cross 1, but all point estimates <1

- Point estimates of HR are significantly higher than other subgroups

Change in slope at~60 months in pts with severe COPD

Survival in COPD patients referred to pulmonologist

Alternative hypothesis:

Pulmonologist HARM patients with COPD as corroborated below:

Referral to Kevin Gibbs

NoYes

Kevin Gibbs and IPF

Kevin Gibbs and IPF treatment

NoYes

*www.facebook.com

Kevin Gibbs and sepsis treatment

NoYes

*www.facebook.com, Levy et al. ANN INTERN MED 2008;148:801-809

Kevin Gibbs and sepsis

COPD exacerbation and β-blocker use

All CI < 1 and point estimates are similar

COPD exacerbations sub group analysis

All CI < 1 except for one

-smaller % of COPD exacerbations in pts with no overt CV disease and incident cases of COPD indicating less severe disease burden vs. those referred to pulmonologist

Discussion Point #1:

Confounding Issues

hypertension

CAD/IHD

A. fib/flutter

primary CMCHF

Pulm HTN

COPD

asthmaSmoking

Discussion Point #2:

What historical factors may have influenced treatment during the study period?

• Prevailing theories on beta blocker use• Available medications

Discussion Point #3:

Would this change your practice?

Why or why not?

Discussion Point #4:

What else do you want to know? What would you hope to see addressed in a clinical trial?