Predictors of oxygen saturation ≤95% in a cross-sectional population based survey

Pulse oximetry has become an important tool in evaluating, and monitoring pulmonary diseases, but the knowledge about arterial oxygen saturation (SpO(2)) values in a general population is limited. The aim of this study was to describe the distribution of low pulse oximetry values in a general adult population and their association with certain predictors.

A cross-sectional population based survey was performed in Tromsø, Norway, in 2007-2008. Valid SpO(2) and pre-bronchodilator spirometry was recorded in 6317 participants aged 38-87 years (57% women). Resting SpO(2) ≤95% was defined as an abnormal value. Predictors of SpO(2) ≤95% with p < 0.05 were entered into a binary multivariable logistic regression.

The prevalence of SpO(2) ≤95% was 6.3%. The strongest predictors in the logistic regression were increased body mass index (BMI) (OR of BMI ≥35 = 6.2, CI(4.2, 9.2)), and reduced forced expiratory volume in 1 s (FEV(1)) % predicted (OR of FEV1% predicted <50 = 4.1, CI(2.5, 6.7)), followed by increased age, male gender, and smoking (p < 0.001). Other significant predictors were elevated C-reactive protein (CRP) (p < 0.005), former smoking, breathlessness and elevated hemoglobin (p < 0.01).

Airflow limitation and other known signs of pulmonary diseases were, as expected, strongly associated with decreased oxygen saturation. Obesity was also a strong predictor of low SpO(2), and should be included among possible explanations when low SpO(2) values are encountered.