The role of Beta-Blockers in Outcomes of Patients Hospitalized for Heart Failure and Chronic Obstructive Pulmonary Disease

This article refers to ‘Outcomes in patients hospitalized for heart failure and chronic obstructive pulmonary disease: differences in clinical profile and treatment between 2002 and 2009’, by L. Staszewsky et al., published in this issue on pages 840–848. The importance of the co-existence of heart failure (HF) and COPD nowadays is widely accepted, and most physicians know that the 2012 ESC guidelines recommend beta-blockers in these patients with both diseases.1 An important conclusion of the study by Staszewsky et al. in this issue of the journal is that the prescription rate of beta-blockers significantly increased from 2002 to 2009.2 In those with HF without COPD, bisoprolol uptake increased from 1.4% to 51.2%, and that of carvedilol from 49.6% to 61.7%. In those with HF and COPD, the increase was from 0.4% to 46.4% for bisoprolol, and from 25.0% to 53.8% for carvedilol. This was one of the major changes in patient management between the cohorts from 2002 and 2009 derived from the same hospitals in Italy. The authors suggest that the improved uptake of beta-blockers could be one of the major reasons for a better prognosis of these HF patients, although they realize it was an observational study.2 Such real-life practice studies are important to confirm results from large randomized clinical trials (RCTs). Certainly for beta-blockers in patients with both HF and COPD, this is of utmost importance because the large landmark RCTs with beta-blockers in HF were selectively in those with a reduced EF (with the exception of the SENIORS trial), excluding patients known to have COPD.3-6 Therefore, RCT data on beta-blockers in patients with both HF and COPD are very scarce. There are also other interesting findings in this study of Staszewsky et al. Because the authors provide data on all types of HF, we may suspect that patients with HF and a normal EF may also benefit from beta-blockade, similarly as suggested in the SENIORS trial.7 Even more important is the finding that all-cause mortality was also significantly lower in those on beta-blockade if AF was also present.2 Comparing the 2009 cohort with the 2002 cohort, all-cause mortality was reduced in those with HF and COPD [adjusted hazard ratio (HR) 0.79, 95% confidence interval (CI) 0.74–0.84], and also in those with HF without COPD (adjusted HR 0.74, 95% CI 0.64–0.84). Similarly, all-cause mortality was reduced in the subgroup of patients with AF (adjusted HR 0.77, 95% CI 0.69–0.87 in those with HF without COPD, and adjusted HR 0.66, 95% CI 0.53–0.81 in those with HF and COPD).2 This is a major contrasting finding to recent meta-analyses of large beta-blocker trials in HF showing that those who also have AF seem not to benefit from beta-blockade.8, 9 An explanation of this contrasting finding could be a difference in dosage of beta-blockers between RCTs and real-life studies. In the landmark HF trials, beta-blocker dosage was titrated to targeted maximal dosage, in this case irrespective of co-existing AF.3-6 Although we do not know details of the dosages of beta-blockers in the study of Staszewsky et al., they were likely to be lower. We know that the dosage of beta-blockade and heart rate are important in HF.10 It could well be that patients with HF and AF should receive lower dosages of beta-blockers than patients with HF but without AF, because too low heart rates can easily hamper cardiac output in such patients. Further research is certainly needed in patients with HF and AF, with a stratified analysis for dosage and heart rate before firm conclusions may be drawn. The diagnosis of COPD in the study of Staszewsky et al. was based on the ICD-9 (International Classification of Disease, Ninth Revision) code and not on results from pulmonary function testing. This does not affect their main findings because this was the case in both the 2002 and the 2009 cohort. Even the use of pulmonary function testing, notably spirometry, may result in overdiagnosing COPD in patients with HF.11, 12 This is because pulmonary fluid overload in HF may cause obstructive alterations by compressing the bronchae from the outside, resulting in obstruction (a reduced FEV1/FVC ratio <0.70) with spirometry.11, 12 Thus, spirometry should be performed when HF patients are stable and euvolaemic.11, 12 Considering the prescription rates of beta-blockade in the study of Staszewsky et al. (2009: 39.4% in those with HF, and 34.5% in HF and COPD), one has to realize that in acute HF beta-blocker treatment may have been withdrawn temporarily. An older study that recruited patients with HF hospitalized for acute cardiac decompensation reported even lower prescription frequencies for beta-blockers (29%).13 The effects of bronchodilators in patients with COPD and HF are less well investigated so far. In the study of Staszewsky et al., those on beta-agonists tended to have a non-significantly worse prognosis, but more than half of the patients with HF and COPD were not treated with these drugs.2 Bronchodilators are known to ameliorate symptom burden in COPD, but without affecting prognosis.14 In HF, multiple studies report negative effects of bronchodilators on prognostic outcomes.15 Beta-agonists are suspected because they increase heart rate and arrhythmia burden and can thus cause a deterioration in LV function.5 In conclusion, the study of Staszewsky et al. provides some important novel data on HF and COPD. Especially important is that in their observational study it is suggested that beta-blockers are prognostically beneficial in the subgroup of HF patients with AF, with or without COPD. Nevertheless, there are many pitfalls when studying co-existing HF and COPD, and some important questions regarding diagnosis and treatment are still not completely resolved. Conflict of interest: none declared.