End-tidal carbon dioxide in the diagnosis of acute pulmonary embolism in hospitalized adult patients

Abstract

Pulmonary embolism (PE) causes 100,000 deaths and contributes to over 4 billion dollars of annual healthcare costs. Acute PE is a diagnostic challenge as symptoms vary widely and are often nonspecific. Definitive diagnosis requires costly confirmatory testing with computed tomography pulmonary angiogram (CTPA). CTPA risks include allergic reactions, contrast-induced nephropathy and radiation exposure. The purpose of this study is two-fold: 1) to examine the accuracy, financial impact, and harm avoidance of adding EtCO[2] to the PE diagnostic process and 2) to evaluate clinician adherence to national guidelines (BPA) for PE diagnosis. A single center prospective, descriptive, correlational design comparing EtCO[2] values to CTPA results. Medical records were reviewed to determine BPA adherence. A total of 111 patients had definitive CT results. Mean ([plus or minus]SD) EtCO[2] was higher for PE+ (28[plus or minus]7.8) versus PE- (33[plus or minus] 8.1) patients (p =.01). For PE exclusion, an EtCO[2] cutoff value of [greater than or equal to]42mmHg yielded a sensitivity of 100%, specificity of 12.2% and a negative predictive value of 100% and could safely eliminate 11 patients (9.9%) from receiving CTPA. For every 6 patients assessed with EtCO[2], one can be saved from CTPA. Overall BPA adherence was 0%. Partial adherence was observed with clinician recorded clinical decisions rules in 3.6% (4/111) and D-dimer was obtained in 10.2% (9/88). EtCO[2] cutoff value of [greater than or equal to]42mmHg could decrease CTPA scans use in [about]10% of adult inpatients suspected of PE eliminating exposure to CTPA risks. Lack of clinical decision support may contribute to low BPA compliance.