Chronic Obstructive Pulmonary Disease: A Slowly Progressive Cardiovascular Disease Masked by its Pulmonary Effects?

This editorial refers to ‘Airflow obstruction, lung function, and risk of incident heart failure: the Atherosclerotic Risk in Communities (ARIC) study’ by S. Agarwal et al., doi. 1093/eurjhf/hfs016 and ‘Clinical characteristics and outcomes of hospitalized heart failure patients with systolic dysfunction and chronic obstructive pulmonary disease: findings from OPTIMIZE-HF’ by R.L. Mentz et al., doi. 1093/eurjhf/hfs009. Until 2003, heart failure was considered uncommon in patients with chronic obstructive pulmonary disease (COPD), and, whenever present, heart failure would occur as right-sided heart failure (cor pulmonale).1 Importantly, early studies included small samples of relatively young patients (mean age 53–68 years), mostly with severe COPD and, unfortunately, selectively excluded patients known to have coronary artery disease. Left ventricular dysfunction [left ventricular ejection fraction (LVEF) <40–50%] was present in 0–16% of these patients.2 Although five studies published between 1975 and 1984 reported much higher prevalence rates (10–46%) in unselected patients (i.e. without excluding patients with coronary artery disease), the general idea that concomitant heart failure was rare in patients with COPD persisted until 2003.2 Then, McCullough et al. showed that 21% of 417 patients with self-reported COPD or asthma (mean age 62 years) presenting at the emergency department with acute shortness of breath had previously unknown heart failure.3 In 2005, similar results were reported among 405 elderly patients (mean age 73 years) with stable COPD.4 While in COPD the abnormalities in the lungs, including pulmonary arterial hypertension, and pulmonary-induced systemic effects such as hypoxaemia, oxidative stress, chronic progressive muscle wasting, and increased sympathetic tone, may induce (right-sided) heart failure, the concomitant presence of COPD and heart failure seems primarily attributable to shared cardiovascular risk factors (and not only cigarette smoking!) and (preclinical stages of) cardiovascular disease. Importantly, these are far more prevalent among COPD patients than in age-matched population controls, even independently of smoking behaviour. Consequently, heart failure in unselected COPD patients is left sided in the large majority, because of its known link with cardiovascular risk factors and disease.4,5 This does not exclude that right ventricular structural changes and functional abnormalities may ultimately lead to deterioration of left ventricular function due to interventricular dependence.6 Common risk factors seem to result in partly overlapping disease pathways, involving low grade systemic inflammation, vascular leakage, and even progressive atherosclerosis. Phenotypically, both diseases show considerable variation, but have a progressive course with multiple exacerbations often necessitating hospitalization, and their systemic effects involve multiple other organs. Both diseases are known to have a large impact on the patient's well-being and on national healthcare budgets. Typically, heart failure occurs during the time course of a patient with COPD, and not vice versa, because COPD starts at an earlier age and has substantially lower mortality rates than heart failure. This is nicely illustrated in the longitudinal study of Agarwal et al. published in this issue.7 The two reports in this issue of the journal are examples of the fascinating interaction between the two common diseases.7,8 In a contemporary, large, acute heart failure registry from the USA (mean age 73 years, 62% male), Mentz et al. show that patients with heart failure with reduced LVEF <40% (HF-REF) and a diagnosis of COPD have an increased burden of co-morbidity, longer length of hospitalizations, a reduced use of evidence-based heart failure therapies, and higher in-hospital mortality rates driven by non-cardiovascular mortality than HF-REF patients without COPD.8 The study of Agarwal et al. is a population-based cohort from the USA in which adults aged 45–64 years are followed for 15 years on average. The hazard ratio for incident heart failure increased with descending quartiles of forced expiratory volume in 1 s (FEV1). Adjustments for age, sex, heart disease, smoking, and other cardiovascular risk factors, and even use of beta-blockers and pulmonary medication (such as beta-agonists), did not alter these findings. These results are consistent with other recently published longitudinal studies in community-based populations.9 The study by Agarwal et al. thus strengthens the hypothesis that pulmonary obstruction itself is a major risk factor for heart failure. These two studies provide useful material to elucidate the entanglement of COPD and heart failure further, and give clues that can help to improve (early) diagnosis and treatment in patients with COPD and possible concomitant heart failure in everyday clinical practice. Since the prevalence of heart failure (with or without reduced ejection fraction) of 26% in patients with COPD is about three times higher than in age-matched population controls,4 cardiologists should condider screening for COPD in patients with heart failure. On the other hand, since as many as 80% remain unrecognized in daily practice, pulmonologists should consider screening for heart failure in COPD patients. Signs, symptoms, electrocardiography, chest X-ray, and N-terminal pro brain natriuretic peptide all have independent diagnostic value.10 Echocardiography is also a cornerstone investigation in these patients, with high-quality images in the vast majority of COPD patients.10 Cardiovascular magnetic resonance imaging could be an alternative to echocardiography in patients with unclear views. Although in the study by Mentz et al. concomitant COPD in heart failure patients did not significantly increase in-hospital all-cause mortality, others have provided evidence that newly detected heart failure in community-dwelling elderly COPD patients doubles mortality rates.11 Also in newly diagnosed hospitalized heart failure patients, those with COPD had higher all-cause 5 year mortality.12 Also FEV1, as a percentage of the predicted value, is an independent predictor for all-cause mortality in patients admitted with heart failure.13 Although not extensively investigated, there seems to be no significant difference in prognosis between HF-REF and HF-PEF (heart failure with preserved ejection fraction) patients with co-existing COPD.14 Patients with heart failure and COPD should not be denied cardioselective beta-blockers (e.g. metoprolol, bisoprolol, or nebivolol), although low-dose initiation and gradual up-titration (‘start low, go slow’) seems even more important than in heart failure patients without COPD.15 Only for heart failure patients with more severe COPD has the efficacy of beta-blockade not been extensively studied.15 Non-selective beta-blockers are not recommended because of the higher risk of bronchospasm, especially in the initial phase, although scarce data suggest that carvedilol can be used safely.15 It is well known that beta-blockers initially have negative inotropic effects, while over time gradually displaying their cardiovascular beneficial effects. This seems also to occur in heart failure patients with COPD.15 In addition, short- and long-term effects of beta-agonists may differ. In particular, orally administered beta-agonists in patients with heart failure can demonstrate acute haemodynamic improvements, with an increase in LVEF and cardiac index, and a decrease in pulmonary wedge pressure, and even improve symptoms and exercise tolerance.15 In the longer term, however, beta-agonists exert numerous unfavourable cardiovascular effects, that are possibly even more pronounced in heart failure patients because of the increased chronotropic and inotropic responsiveness to beta-agonists and vulnerability to arrhythmias of the failing myocardium. These negative effects include tachycardia and disturbed autonomic modulation with depressed heart rate variability, and, in susceptible patients, could exert ischaemic events.15 Importantly, standard metred-dose beta-agonist inhalers produce only minor systemic and biochemical abnormalities. Although observational studies reported that beta-agonist inhalers in patients with heart failure and COPD may increase mortality and heart failure hospitalizations, adjustment for confounders showed that purported adverse effects of beta-agonists may be attributable to underlying pulmonary disease.15 In a post-hoc analysis of participants in the CHARM programme, Hawkins et al. show that beta-blockers seem capable of counter-balancing possible negative effects of bronchodilators, including beta-agonists, on survival, although the authors were not able to adjust for the severity of COPD.16 Randomized controlled trials have established the safety of long-acting beta-agonists in patients with COPD; however, concerns remain regarding the safety of these drugs in patients with asthma. The long-acting anticholinergic bronchodilator tiotropium has proven efficacy on symptoms and pulmonary function in both COPD and asthma, is as equally effective as the long-acting beta-agonist salmeterol, and has reassuring cardiovascular safety data. Therefore, tiotroprium is an alternative to beta-agonists. Considering ‘always’ continuing beta-blockers in patients with heart failure and COPD is also underpinned by the potential beneficial pulmonary effects of long-term beta-blocker use, based on animal models. Long-term use of beta-blockers can up-regulate beta2-receptors in the lungs and thus reduce the need for beta-agonists and reduce inflammation.17,18 Interestingly, recent large observational studies provide some evidence that long-term use of beta-blockers may reduce mortality rates and reduce the risk of pulmonary exacerbations in patients with COPD, even in the subgroup of patients not known to have cardiovascular disease, and irrespective of the use of beta-agonist inhalers.19,20 Residual confounding in these studies is likely to dilute the beneficial effect of beta-blockers, because beta-blockers are typically prescribed in patients with increased mortality risks (i.e. with hypertension and cardiovascular disease). Reducing the sympathetic tone and up-regulation of beta2-receptors in the lungs could be possible mechanisms by which beta-blockers exhibit pulmonary beneficial effects in the long term.17,18,21 Thinking of heart failure as a possible cause in any patient with shortness of breath and fatigue, or an increase in such symptoms, irrespective of other disease labels, including COPD, means that physicians need to ‘reset’ their clinical reasoning. This applies to cardiologists as well as non-cardiologists, such as pulmonologists, general practitioners, and geriatricians. Dual diagnosis results in an increased burden of care including complex polypharmacy and multiple providers with potentially competing therapeutic strategies and the risk of misinterpretation of therapy and adverse effects of treatment for one disease on the co-morbid conditions.22 The time has come for physicians to place their trust in beta-blockers in patients with heart failure and COPD. Future randomized trials are needed to confirm the long-term positive results of beta-blockers in patients with COPD, even in those without overt cardiovascular disease. A positive result of such an RCT on important clinical endpoints would cause a paradigm shift comparable with beta-blocker use in patients with heart failure in the 1990s. Conflict of interest: none declared.