The GAS trial Andrew Davidson and colleagues’ GAS trial (Jan 16, p 239)1 reports outcomes that are different to those initially registered on (NCT00756600). There was one prespecified primary outcome, the Wechsler Preschool and Primary Scale of Intelligence (third edition) Full Scale IQ score, measured at 5 years corrected age, which was not reported; but with clear reasons for this stated in the paper. There were three prespecified secondary outcomes. One of which, the Bayley neurodevelopmental scale, was not clearly defined. The other two, frequency of apnoea and characteristics of apnoea, in the postoperative period and measured at 2 years, were not reported in the paper.

Additionally, the paper reports 26 new secondary outcomes that were not prespecified, without declaring them as such. 21 of these non -prespecified outcomes are subtypes of the Bayley score. CH has received grant funding from the WHO and the National Institute for Health Research School of Primary Care Research and is a founder of the AllTrials campaign. ES and HD declare no competing interests. Eirion Slade, Henry Drysdale, *Carl Heneghan, on behalf of the COMPare project team [email protected] Centre for Evidence-Based Medicine, University of Oxford, Oxford OX2 6GG, UK 1 Davison AJ, Disma N, de Graaff JC, et al, for the GAS consortium. Neurodevelopmental outcome at 2 years of age after general anaesthesia and awake-regional anaesthesia in infancy (GAS): an international multicentre, randomised controlled trial. Lancet 2016; 387: 239–50.

The ongoing GAS trial compares cognitive outcomes in infants randomly assigned to receive either awake-regional anaesthesia or sevofl urane-based general anaesthesia during hernia repair. The primary outcome is cognitive performance, measured at 5 years of age. The preliminary report1 described the secondary outcome of cognitive performance at 2 years of age, based on a sample of 532 patients. The average duration of anaesthesia was 54 min. At 2 years of age, equivalent cognitive scores were observed in the two groups. Although the results are welcome, the data must be considered in the Daniel Leal-Olivas/Corbis See Editorial page 1591 Correspondence 1614 Vol 387 April, 16, 2016 proper context. This Correspondence represents a consensus response to the GAS study from SmartTots, a public– private partnership. Researchers who endorsed this Correspondence are listed on the SmartTots website