Hypoxaemia definition in childhood based on arterial oxygen saturation obtained with a pulse oximeter: a retrospective study

Objective Using normal pulse oximetry (SpO 2 ) values as basis, an oxygen saturation <98% could reflect hypoxaemia in childhood. The objectives were to define the level of SpO 2 associated with hypoxaemia, the prevalence of SpO 2 <98% in children suffering from chronic diseases referred to pulmonary function testing units. Setting Two university paediatric hospitals. Design and patients We retrospectively selected arterialised blood gas (ABG) analyses and SpO 2 recordings, and further recorded pulmonary function test (PFT) indices (z-scores of forced expiratory volume at 1 s (FEV 1 ), forced vital capacity (FVC), FEV 1 /FVC and total lung capacity) of visits with SpO 2 <98% (Radical-7 pulse oximeter, Masimo). Results The Bland-Altman analysis of saturations (n=1188) showed that SpO 2 overestimated arterialised saturation (bias: +1.3%, precision: ±1.1%). The 20 ABGs with SpO 2 ≤95% showed hypoxaemia (z-score of arterialised pressure of O 2 <−1.96), while 48/74 (65%) ABGs with SpO 2 of 96% or 97% showed hypoxaemia. Three thousand eight hundred twenty-four PFTs were recorded in 2641 children (1186 girls, 1455 boys) with a mean age (±SD) of 12.3±3.5 years. The prevalence of SpO 2 <98% was 114/2641 children (4.3%, 95% CI 3.6 to 5.2), showing that hypoxaemia was mainly observed in chronic obstructive diseases (96 children). Twelve children had SpO 2 ≤95% and all had ventilatory defects (based on international definitions of PFT). Conclusions An oxygen saturation ≤95%, using the Masimo oximeter, is indicative of hypoxaemia and PFT abnormalities, while oxygen saturation of 96% or 97% indicates possible hypoxaemia. Thus, chronic respiratory insufficiency can confidently be diagnosed if repeated SpO 2 is ≤95% using the Masimo oximeter.