@ARTICLE{Abiri, author = {Abiri, Samaneh and Hatami, Naser and Habibzadeh, Seyed Reza and Mehramiz, Neema John and Yaghoubian, Hossein and Kalani, Navid and Ghaedi, Mojtaba and }, title = {Sensitivity and Specificity of Physical Examination in the Diagnosis of Pneumothorax and Hemothorax}, volume = {9}, number = {1}, abstract ={Introduction: The aim of this study was to evaluate the sensitivity and specificity of history and physical examination in the diagnosis of pneumothorax and hemotorax in blunt chest trauma patients. Methods: This was a descriptive-analytical study. Physical examination results were compared with findings of chest CT, X-ray, and ultrasound diagnostic methods using ROC curves in SPSS software. Findings: Three (3.03) patients with pneumothorax and 7 (7.07) patients with hemotrax were found among 99 patients with mean age of 33.4±19.43. The highest sensitivity was due to chest scraping for pneumothorax (66.67%) and hemothorax (100%). The highest specificity was for abnormal lung sounds (Crackle), with 96.88% specificity for pneumothorax and 98.89% specificity for hemothorax. In the study of pneumothorax, the highest PPV and NPV were related to pulmonary sound reduction (12.5% and 98.7%, respectively). In the hemothorax evaluation, the highest PPV was related to chest tenderness (37.5%) and the highest NPV to pulmonary sound reduction (96.3%). The highest accuracy for pneumothorax was for pulmonary sound reduction and abnormal pulmonary sounds for hemothorax. A heart rate above 98.5 was associated with pneumothorax with a sensitivity of 17.6% and a specificity of 66.7%. Diastolic blood pressure below 70.5 with 46.9% sensitivity and 85.7% specificity and respiratory rate below 6.5 with 92.6% sensitivity and 57.1% specificity were associated with hemothorax. Conclusion: Proper physical examination and history taking can help to diagnose hemothorax and pneumothorax with high sensitivity and specificity complementarity to CT scan or X-ray results. }, URL = {http://intjmi.com/article-1-471-en.html}, eprint = {http://intjmi.com/article-1-471-en.pdf}, journal = {International Journal of Medical Investigation}, doi = {}, year = {2020} }