(1191) Oxygen for Moderately Hypoxemic Children in Malawi: NoGoLO2 Pilot Randomized Controlled Trial

Introduction: World Health Organization (WHO) guidelines recommend oxygen therapy for children with a peripheral oxyhemoglobin saturation (SpO2) < 90% in low- and middle-income countries (LMICs). Although children with pneumonia and a SpO2 of 90-93% in LMICs have an elevated mortality risk it is unknown whether oxygen treatment improves their outcomes. We aimed to determine the feasibility of conducting a randomized controlled trial (RCT) in Malawi comparing: low-flow oxygen or high flow nasal cannula (HFNC) oxygen, to standard of care without oxygen treatment (control), among children with a SpO2 of 90-93%.

Methods: We conducted a pilot open-label, 3-arm RCT called NoGoLO2 at a district hospital in Salima, Malawi in 2024. Eligible children 1-59 months old with WHO-defined pneumonia, an SpO2 90-93% measured by pulse oximetry (Masimo RadG) in room air, and no WHO-defined emergency signs provided written informed consent and were randomized to control, nasal prong low-flow oxygen from an oxygen concentrator, or HFNC oxygen (Airvo3 Optiflow System, Fisher & Paykel). All participants were hospitalized and otherwise received care per WHO guidelines. The primary outcome was feasibility, defined as < 2 protocol deviations per participant. Secondary outcomes included consent refusal, treatment failure within 14 days, and participant attrition (ClinicalTrials.gov registration: NCT06176664).

Results: Of 29 eligible participants, 21 (72%) were consented and randomized (7 per arm). There were 16 protocol violations overall (mean < 1 per participant); 7 (43.8%) were intervention arm violations (3 assessment delays, 4 incorrect oxygen weans) and 9 (56.3%) were from first-line medication stock-outs requiring alternative regimens. No participants withdrew and 3 (14.3%) had treatment failure (1 HFNC arm, 2 control arm (1 death), and 0 Low-Flow arm). There were no missing 14-day outcomes, including 16/16 successful phone encounters, 4/4 home tracing visits for those lacking phone access, and 1 in-hospital death.

Conclusions: A three-armed, open-label randomized controlled trial comparing low flow and HFNC oxygen to standard care among children with a SpO2 90-93% in Malawi was feasible. Lessons from this pilot trial will inform the development of the full NoGoLO2 trial.