Intravenous salbutamol in ARDS and increased mortality – Authors' reply

Fekri Abroug and colleagues note significant divergence between predicted and actual mortality in the BALTI-2 trial.1Gao Smith F Perkins GD Gates S et al.Effect of intravenous beta-2 agonist treatment on clinical outcomes in acute respiratory distress syndrome (BALTI-2): a multicentre, randomised controlled trial.Lancet. 2012; 379: 229-235Summary Full Text Full Text PDF PubMed Scopus (247) Google Scholar Our predicted mortality was based on the Acute Physiology and Chronic Health Evaluation (APACHE II) model. UK coefficients for the model were published in 1990, but recent studies have shown that, even with sophisticated recalibration, the model fit is far from optimal.2Harrison DA Brady AR Parry GJ et al.Recalibration of risk prediction models in a large multicenter cohort of admissions to adult, general critical care units in the United Kingdom.Crit Care Med. 2006; 34: 1378-1388Crossref PubMed Scopus (136) Google Scholar In APACHE II, the standardised mortality ratio is referred to as a ratio of actual to predicted hospital mortality (rather than 28-day mortality).3Rowan K Risk adjustment for intensive care outcomes.in: Goldhill DR Withington PS Textbook of intensive care. 1st edn. Chapman & Hall, London1997: 787-798Google Scholar As shown in our original table 2,1Gao Smith F Perkins GD Gates S et al.Effect of intravenous beta-2 agonist treatment on clinical outcomes in acute respiratory distress syndrome (BALTI-2): a multicentre, randomised controlled trial.Lancet. 2012; 379: 229-235Summary Full Text Full Text PDF PubMed Scopus (247) Google Scholar the actual hospital mortality was 39% in the salbutamol group and 33% in the placebo group. The predicted hospital mortality from APACHE II for salbutamol and placebo was listed in table 1 (0·43 vs 0·42, respectively). So in this cohort, the standardised mortality ratio should be 0·91 for the salbutamol group and 0·79 for the placebo group, rather than 0·79 versus 0·54 as calculated by Abroug and colleagues. We based the sample size calculation on the following information. According to unpublished data for 2005 from the Intensive Care National Audit and Research Centre, the hospital mortality for 37 726 patients with severe respiratory failure in the UK was 41·2%. The primary outcome for BALTI-2 was 28-day mortality, which might be similar to or slightly higher than hospital mortality because most deaths would occur in the intensive care unit within a short period after randomisation, and most patients would leave hospital before 28 days. In BALTI,4Perkins GD McAuley DF Thickett DR Gao F The beta-agonist lung injury trial (BALTI): a randomized placebo-controlled clinical trial.Am J Respir Crit Care Med. 2006; 173: 281-287Crossref PubMed Scopus (378) Google Scholar the placebo group 28-day mortality rate was 66% (95% CI 0·45–0·83). A reasonable conservative estimate of the expected 28-day mortality in BALTI-2 was 40–50%. Mortality in the placebo group of BALTI-2 was much lower than expected. Mortality in acute respiratory distress syndrome (ARDS) has been falling consistently over the past decade, which could be associated with improvement in ARDS treatment. The 28-day mortality reported in our study is similar to that of recent ARDS studies internationally—eg, Papazian and colleagues5Papazian L Forel JM Gacouin A et al.Neuromuscular blockers in early acute respiratory disdress syndrome.N Engl J Med. 2010; 363: 1107-1116Crossref PubMed Scopus (1587) Google Scholar reported 28-day mortality of 33% (95% CI 26·5–40·9). In our study, the survival curves show clearly that salbutamol is not beneficial. The difference between the groups is inevitably smaller at earlier time points because of the smaller number of deaths, and the fact that the curves continue to diverge up to 28 days suggests that salbutamol might have complex effects that persist beyond the period of its administration. FGS and GDP have received an investigator-led research grant from GlaxoSmithKline. GDP has consulted for, sat on advisory boards for, and received lecture fees from GlaxoSmithKline, and has received lecture fees from AstraZeneca for educational meetings (all unrelated to β agonists). The other authors declare that they have no conflicts of interest. Intravenous salbutamol in ARDS and increased mortalityIn the BALTI-2 trial, Fang Gao Smith and colleagues (Jan 21, p 229)1 report an increase in 28-day mortality after a 7-day infusion of salbutamol compared with placebo in patients with early acute respiratory distress syndrome (ARDS). Major discrepancies between predicted and observed mortality make interpretation of this trial difficult, especially when these figures are compared with those reported in Smith and colleagues' previous study (BALTI),2 which was used to calculate the sample size. Full-Text PDF BiDil: alive and kicking – Author' replyLaurence Downey makes four points in his letter in response to my Jan 14 essay. Full-Text PDF