O7.2.2 Safety of beta-blockers

Beta blockers have well established survival benefits in heart failure and after myocardial infarction and have been long used in coronary artery disease and hypertension but have been considered contra-indicated in patients with COPD. A Cochrane systematic review identified 20 RCTs of cardio-selective beta blockers which examined lung function and respiratory symptoms in 278 patients with COPD (Salpeter 2005Salpeter 2002) [evidence level I]. Eleven studies were of single dose and nine were of prolonged treatment (mean 3.7 weeks, range two days to 12 weeks). The beta blockers included atenolol, metoprolol, bisoprolol, practolol, celiprolol and acebutolol and were used at therapeutic doses. There was no significant overall change in FEV1, increase in respiratory symptoms or change in the response to inhaled beta2 agonists. The authors concluded that cardio-selective beta blockers were safe and should not be withheld, even in patients with severe airflow limitation. However, even with pooled data the absolute patient numbers were small and failed to represent minority groups such as females and the elderly. The longest duration included trial was 12 weeks, and so the meta-analysis provides little guidance about long term safety and potential morbidity of prolonged beta-blocker use in COPD. COPD symptoms are more pronounced during exertion and hence a recent study (Mainguy 2012)[evidence level II] has investigated dynamic lung function in GOLD stage II-III COPD patients during cardioselective beta-blocker treatment. This randomised, double blinded, placebo-controlled cross over trial compared inspiratory capacity (IC) at peak isotime exercise during endurance exercise testing as a measure of dynamic hyperinflation for treatment arms with bisoprolol (titrated to 10 mg) and titrated placebo. Included patients had no recognised indication for beta-blocker treatment and were randomised to treatment sequence. As expected, IC reduced with exercise; this effect was more pronounced in the bisoprolol arm -0.50 versus -0.41 (p<0.01). Exercise duration was non-significantly reduced in the bisoprolol arm, and the change was strongly correlated with the change in IC. However, the absolute difference was modest and of arguable clinical significance and so the groups recommended that cardioselective beta-blockers not be withheld on this basis. Beta-blockers have duration-dependent effects, including effects on beta-receptor regulation, which are important in their efficacy for heart failure and may also be relevant in the airways. The 14-day beta-blocker treatment duration in this study was likely insufficient to fully demonstrate these effects. Despite a paucity of evidence to suggest harm, beta-blockers are still under-utilised in COPD for guideline-based indications such as systolic heart failure (Lipworth 2016) [evidence level III-2]. Australian data from a COPD cohort hospitalised for a COPD exacerbation also reflects this (Neef 2016) [evidence level III-2].

A number of observational studies also lend confidence to beta-blocker prescribing in COPD patients. Du et al’s meta-analysis (Du 2016) of 15 observational cohort studies with follow up ranging from one to 7.2 years suggested that beta-blocker treatment was associated with reduced mortality risk (RR 0.72, 95% CI 0.63-0.83) and exacerbation risk (RR 0.63, 95% CI 0.57-0.71). Although there was significant heterogeneity amongst the included studies, sensitivity analysis to account for this did not change the outcome [evidence level III-2].  Beta-blocker treatment did not diminish the beneficial effects of inhaled treatments on post bronchodilator FEV1 or COPD exacerbations. Prospective randomised controlled data is still needed (Dransfield 2018).

Acknowledging that, beyond improving survival, treatments used in COPD patients must not impair QoL, van Gestel’s 2009 study (van Gestel 2009) [evidence level III-2] sought reasons for suboptimal beta-blocker prescription in terms of QoL. The surviving 469 COPD patients of their vascular surgery cohort were subsequently assessed via health-related QoL questionnaire (SF-36) at median follow up of 6.4 (2.9 to 9.3) years. A 70% response rate was achieved. Although 71% of patients were receiving beta-blockers at follow up, compared to 59% at baseline, neither beta-blocker treatment at baseline, nor at follow up, impacted significantly on QoL scores.