Optimising ventilation in preterms with closed-loop oxygen control
Research type
Research Study
Full title
Does closed-loop automated oxygen control reduce the duration of mechanical ventilation? A randomised controlled trial in ventilated preterm infants
IRAS ID
297749
Contact name
Anne Greenough
Contact email
Sponsor organisation
King's College London
Duration of Study in the UK
2 years, 0 months, 0 days
Research summary
Research Summary:
Many premature infants require respiratory support in the newborn period. Mechanical ventilation although life-saving is linked to complications for the lungs and other organs and its duration should be kept to a minimum. The use of supplemental oxygen may also increase the risk of comorbidities such as retinopathy of prematurity. Therefore, oxygen saturation levels and the amount of inspired oxygen concentration provided should be continuously monitored.
Oxygen control can be performed manually or with the use of a computer software incorporated into the ventilator that is called ‘closed loop automated oxygen control’(CLAC). The software uses an algorithm that automatically adjusts the amount of inspired oxygen to maintain oxygen saturation levels in a target range. Evidence suggests that CLAC increases the time spent in the desired oxygen target range but there are no data to determine the effect on important clinical outcomes. A previous study has also demonstrated that CLAC reduces the inspired oxygen concentration more rapidly when compared to manual control. That could help infants come off the ventilator sooner.
With this study we want to compare the time preterm infants spend on the ventilator when we use the software to automatically monitor their oxygen levels with those infants whose oxygen is adjusted manually by the clinical team. That could help us understand if the use of automated oxygen control reduces the duration of mechanical ventilation and subsequently the complications related to it.Summary of Results:
We recruited 70 patients following informed parental consent. One infant was subsequently excluded due to underlying cytomegalovirus infection that was diagnosed after randomisation. There were 34 infants in the intervention group (automated oxygen control) and 35 in the control group (manual oxygen control). The 69 infants were born at a median gestational age of 27.0 (interquartile range (IQR) 25.6-29.0) weeks and had a median birth weight of 795 (650-1105) grams, and were studied at a median corrected postmenstrual age of 27.6 (25.9-29.1) weeks.
The two groups had comparable ventilatory requirements at the beginning of the study and at the time of first extubation attempt.
Automated oxygen control significantly reduced the median percentage of time spent below the target oxygen saturation range (SpO2<91%; p= 0.004) and in hypoxemia (SpO2< 85%; p=0.001), but the median concentration of inspired oxygen that was delivered during hypoxemia was not significantly higher (p= 0.159). Infants in the intervention group spent a greater proportion of their time within the target oxygen saturation range (SpO2: 91%-95%; p=0.001) and the median oxygen requirement in target range did not differ significantly between the groups (p=0.201). The time spent in hyperoxemia (SpO2>95%) was significantly lower during automated oxygen control (p=0.001) and the median oxygen delivery during hyperoxemia was lower (p=0.014).
Automated oxygen control significantly reduced the median duration of mechanical ventilation (median (range): 11 (1-57) versus 40 (3-134) days, p=0.027), and median duration of supplemental oxygen treatment (median (range): 33 (0-100) versus 47 (3-335) days , p=0.031) and median time to first extubation attempt (median (range): 4 (1-52) versus 12 (2-107), p= 0.018) with similar extubation success rates (23% in both groups, p=1.0). The incidences of bronchopulmonary dysplasia (BPD) (55% versus 83.9% , p=0.015) and the combined outcome of moderate/ severe BPD (41.2% versus 74.3%) or death (14.7% versus 11.4%) at 36 weeks postmenstrual age were lower (p=0.008). In addition, fewer infants in the intervention group were discharged on home oxygen (26.5% versus 51.4%, p=0.016). The median length of neonatal unit stay (p= 0.205) and the incidence of retinopathy of prematurity requiring treatment (p=0.168) did not differ significantly between the two groups.REC name
Yorkshire & The Humber - Sheffield Research Ethics Committee
REC reference
21/YH/0139
Date of REC Opinion
16 Jul 2021
REC opinion
Further Information Favourable Opinion