There a few few remarkable things about this study; the first is part of the preamble:
Since 2008, all CDH patients born in European countries represented in the CDH EURO consortium have been treated according to a standardized neonatal treatment protocol that was developed at a consensus meeting. After implementation of this protocol, mortality decreased from 33% to 12%. This decrease in mortality rate should be interpreted against of the year upon year variability in mortality rates. Nevertheless, standardization of care is an ideal backcloth to undertake multicenter randomized controlled trials (RCTs).
In this study full term babies with an antenatal diagnosis of CDH were randomized to start on either conventional ventilation (Initial settings were a positive inspiratory pressure (PIP) of 20 to 25 cmH2O and a positive end-expiratory pressure (PEEP) of 3 to 5 cmH2O, with a ventilator rate of 40 to 60/min) or high frequency ventilation (mean airway pressure 13 to 17 cmH2O, frequency 10 to 12 Hz, delta P 30 to 50 cmH2O depending on chest wall vibration). There were a fairly complex list of signs of failure of the initial ventilation mode (If these things happen for at least 3 hours: inability to maintain preductal saturations above 85% or postductal saturations above 70% ; increase in CO2 > 65 mm Hg despite optimization of ventilatory management; PIP > 28 cmH2O; mean airway pressure >17 cmH2O; inadequate oxygen delivery with metabolic acidosis defined as lactate 5 mmol/L and pH < 7.20; hypotension resistant to fluid therapy and inotropic support resulting in a urine output <0.5 ml/kg/hour; oxygenation index of 40). In which case they could be switched to the other ventilation mode, and if that didn’t work they could be placed on ECMO.
The primary outcome of the study was survival without chronic lung disease (oxygen requirement at 28 days). Over 5 years they enrolled 171 patients, far short of the hoped for 200 per group. In the Conventional group 45% died or had chronic lung disease compared to 53% in the HFO group, a difference which may have been due to chance, (p= 0.3). Many of the secondary outcomes were better in the conventional group, there was much less need for ECMO (26% vs 51%), they were ventilated for fewer days,
less often received inhaled nitric oxide, or sildenafil, had a shorter duration of
vasoactive drugs, and less often failed treatment.
I think the study is a remarkable achievement, as far as I can see this is by far the largest RCT of infants with CDH that has ever been performed. Even though it was unfortunately stopped before the full sample size, it looks unlikely that an advantage of initial HFO would have been shown. As we are unlikely to have any other data of this quality for a very long time, I think the conclusion should be that the starting assisted ventilation mode for infants with CDH should be conventional ventilation, unless there is some very good reason for choosing HFO.
We can only hope that this collaborative group will continue to perform important clinical studies in these babies, who remain a group with very high mortality despite the advances of the last several years. I would vote for a good trial of sildenafil, a comparison of different criteria for ECMO, and more studies of different aspects of assisted ventilation. Maybe a study of muscle relaxation as well, if that wasn’t asking too much!