Revista Educação em Saúde (Sep 2018)
Performance da CO-oximetria de pulso para identificação do tabagismo na Atenção Básica
Abstract
Objective: To evaluate the performance of a CO-oximeter to determine the smoking status in relation to an exhaled carbon monoxide meter in subjects attending Primary Health Care. Methods: A cross-sectional study was carried out in outpatients aged 18 years or older, attending Primary Health Care Units of Anápolis-GO. Subjects were submitted to clinical evaluation, exhaled carbon monoxide dosing, and carbon monoxide measurement by pulse CO-oximetry. Results: A total of 449 individuals were included, 106 smokers (23.6%), 89 ex-smokers (19.8%), 203 never smokers (45.2%) and 51 passive smokers (11.4%). CO-oximetry with cutoff point of ≥4% showed a specificity of 77.3% (95% CI 72.5% -81.6%), a sensitivity of 52.8% (CI 95% 42.9% - 62,6%), positive and negative predictive values of 45.3% (95% CI: 38.9% -51.9%), 82.1% (95% CI: 78.8% -85%), respectively and area under ROC curve of 0.73 (95% CI 0.67-0.78). The exhaled carbon monoxide dosage with cut-off point of ≥5ppm showed a specificity of 96.2% (95% CI 93.6% -98%), sensitivity of 79.2% (95% CI 70.3% -86, 5%), positive and negative predictive value of 88.2% (95% CI 81.3% -92.8%), 92.9% (95% CI 89.9% -95%), respectively, and area under ROC curve of 0.91 (95% CI 0.87-0.95). COHb(%) ≥4 showed OR: 4.3; (95%CI 2.7-7.1); p <0.0001. Conclusion: CO-oximetry showed a lower capacity to detect smoking than the exhaled carbon monoxide measure. However, when COHb (%) ≥4, there is a high probability of smoking.
Keywords