Background: Several studies have reported that computed tomography pulmonary angiography is the best method for diagnosing pulmonary embolism (PE). This study, however, aimed to predict or exclude PE using the end-tidal carbon dioxide (ETCO2) value and alveolar dead space fraction (AVDSf) together. Methods and Materials: One-hundred patients were included in the present study. Patients with suspected PE were evaluated using clinical prediction rules proposed by the Wells and the Modified Geneva scoring systems. PE was ruled out in patients with normal D-dimer concentrations (< 0.55 mg/dl). Patient ETCO2 values were recorded using time versus waveform capnography before performing imaging studies. Capnography was performed for 2 min; however, the average ETCO2 values measured over the final 1 min were recorded in "full continuous" mode. Arterial puncture was performed simultaneously for arterial blood gas analysis. Additionally, AVDSf was calculated according to the Bohr equation. Results: PE was detected in 36 % of patients. Patients were classified into high-, moderate, and low-risk groups according to the Wells and Modified Geneva scores. PE was excluded in 95 % and 100 % of patients with low Wells and Modified Geneva system scores, respectively, when ETCO2 was > 28.5 mmHg. The diagnosis of PE was excluded in 100 % of patients with low Wells and Modified Geneva scoring system scores with AVDSf < 0.128. High wells and Modified Geneva system scores were helpful in diagnosing of PE (100%) when AVDSf was > 0.128. Conclusion: It was possible to exclude/predict PE based on ETCO2 and AVDSf values calculated using capnography when evaluated with clinical prediction rules and D-dimer test using an a lgori thm. 0Z019 Sociedade Portuguesa de Pneumologia. Published by Elsevier Espana, S.L.U. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).