Abstract
Background: Pulse oximetry has become an important tool in evaluating, and monitoring pulmonary diseases. There is limited knowledge on distribution of SpO2 values in a general population.
Aims and objectives: We wanted to determine independent predictors of low pulse oximetry values in a screened population.
Methods: A cross-sectional population based survey was performed in the city of Tromsø, Norway, in 2007-2008. Valid spirometry and pulse oximetry (SpO2) were performed in 6320 participants aged 38-87 years (57% women). The examinations also included questionnaires and readings for hemoglobin, C-reactive protein (CRP), pulse, weight and height. We considered resting oxygen saturation ≤ 95 as an abnormal value. Predictors of SpO2 ≤ 95% with a statistical significance of p<0.25 were entered into a logistic regression model. In the final model we included predictors with p≤0.05.
Results: We found SpO2 ≤ 95% in 400 participants (6.3%) and SpO2 ≤ 92 in 30 (0.5%). The strongest predictors in the logistic regression (p<0.001) were increased BMI, reduced FEV1% predicted, and increased age, hemoglobin and CRP. Other significant predictors were current smoking (p=0.001) and former smoking (p=0.037). The following variables did not reach statistical significance; pulse, self-reported asthma, COPD, cardiovascular disease, atrial fibrillation, diabetes and hypertension, neither did pack-years, dyspnea, cough, and recent airway infection. Sex was an insignificant factor in the final model.
Conclusion: Independent predictors of SpO2 ≤ 95% in this population based survey were BMI (↑), FEV1% predicted (↓), age (↑), hemoglobin (↑), CRP (↑), and history of smoking (↑).
- © 2011 ERS