Potential risk of hypoxaemia in patients with severe pneumonia but no hypoxaemia on initial assessment: a prospective pilot trial

Abstract

Background: The World Health Organization recommends oxygen therapy for children under 5 years of age with pneumonia and lower chest indrawing. In patients with severe pneumonia who are initially normoxaemic, there is little information on the risk of subsequently developing hypoxaemia and the benefit of routine oxygen therapy.Objectives: To study the incidence of subsequent hypoxaemia in initially normoxaemic children with pneumonia and lower chest indrawing.Methods: Children (n = 58, 3–59 mths) with pneumonia, lower chest indrawing and normoxaemia (SpO2 >90%) were randomly assigned to receive supplemental oxygen (nasal prongs, 1–2 L/min flow) (n = 29) or room air (n = 29). Vital signs and SpO2 were monitored continuously and recorded every 6 hours. Outcome variables were incidence of hypoxaemia, length of tachypnoea and lower chest indrawing.Results: The two groups had similar demographic and clinical profiles. Thirty-one patients (53%) developed hypoxaemia later, without significant differences between the two arms (RR 0·61, 95% CI 0·36–1·04). Patients who developed hypoxaemia later were similar to those who did not, except for a lower SpO2 on enrolment. However, they took more time to recover from tachypnoea (P<0·05), chest indrawing (P<0·05) and fever, indicating that they had more severe disease. Early oxygen therapy did not alter the course of disease.Conclusions: About half of the normoxaemic patients with severe pneumonia developed hypoxaemia after enrolment, indicating a significant potential risk. Children hospitaled with severe pneumonia might benefit from routine oxygen therapy. Alternatively, oxygen might be provided to those who develop hypoxaemia identified by a pulse oximeter.