Accuracy of carbon monoxide in venous blood to detect smoking in male and female trauma patients

Abstract

Background: The determination of the smoking status should be part of the initial assessment in emergency or surgery patients, as preventive measures might reduce long-term morbidity. Next to self-report (not always possible or reliable) the use of biomarkers such as carbon monoxide (CO) has been recommended. Point-of-care CO-hemoglobin (CO-Hb) determination is routinely used in trauma settings. The aim of this study was to evaluate the accuracy and the best cut-off of CO-Hb in order to determine the smoking status in minimally injured patients in the emergency department. Methods: After informed consent and ethical committee approval 1283 minimally injured patients were assessed in the emergency department immediately after admission with a computer-based questionnaire including the ‘Heaviness of Smoking Index’ questionnaire. Venous blood sampling and point-of-care oximetry, including CO-Hb, was performed. Results: Median age of all patients (smokers: males: 399, females: 176; nonsmokers: males: 420, females: 288) was 33 years. According to the receiver operating characteristics CO-Hb had an excellent ability to differentiate between smokers and nonsmokers (area under the curve: males: 0.949, females 0.932). Optimal CO-Hb cut-off was 1.8% in males (specificity 95.7%, sensitivity 85.5%) and 1.6% in females (specificity 96.9%, sensitivity 80.1%). Conclusion: CO-Hb was a good marker to detect current smoking in trauma patients. The cut-off of CO-Hb should be lowered to 1.6% in females and 1.8% in males.