Computed Tomographic Pulmonary Angiographic Findings Can Predict Short-Term Mortality of Saddle Pulmonary Embolism: A Retrospective Multicenter Study. Issue 3 (May 2016)

• Record Type: Journal Article
• Authors: Liu, Min; Guo, Xiaojuan; Zhu, Li; Zhang, Hongxia; Hou, Qing; Guo, Youmin; Yang, Yuanhua; Jiang, Tao
• Is Part Of: Journal of computer assisted tomography. Volume 40:Issue 3(2016)
• Publication Date: 2016-05
• Subjects: saddle pulmonary embolism -- computer tomographic pulmonary angiography -- right ventricular dysfunction -- short-term mortality; Tomography -- Periodicals; Tomography -- Periodicals; Tomography; Periodicals; 616.0757
• DOI: 10.1097/RCT.0000000000000373
• Languages: English
• ISSNs: 0363-8715

Abstract

Abstract : Objective: In patients with saddle pulmonary embolism (PE), the correlation between computed tomographic pulmonary angiographic (CTPA) findings and short-term outcome remains unclear. The purpose is to determine if CTPA findings predict 1-month mortality of patients with saddle PE. Methods: This is a multicenter, retrospective study of saddle PE. Computed tomographic pulmonary angiographic findings of 115 consecutive patients (male-to-female ratio, 65:50; mean age, 64.3 ± 16.3 years) with saddle PE were evaluated. One-month mortality after diagnosis was the primary end point. Results: Twenty-four patients died within 1 month. Among CTPA findings, quantitative parameters including right/left ventricular area ratios (RVa/LVa), right/left atrial diameter ratios, Cobb angle, and Mastora score were significantly enlarged in survivors. Also, qualitative findings including contrast agent reflux into the azygos vein and pericardial effusion were significantly different between survivors and nonsurvivors. Areas under the curve on receiver operating characteristic curves revealed the cutoff values for predicting early mortality of saddle PE using right/left atrial diameter ratios, RVa/LVa, Mastora score, and Cobb angle, respectively, were 2.15, 2.00, 69%, and 58°. Logistic regression analysis suggested that both RVa/LVa (odds ratio, 5.100; P = 0.0004) and Cobb angle (odds ratio, 1.596; P = 0.0321) were independent predictors of early mortality. The combination of RVa/LVa and Cobb angle increased the area under the curve to 0.882, but the difference did not reach significance compared with RVa/LVa or Cobb angle, alone ( P > 0.05). Conclusion: In patients with saddle PE, RVa/LVa and Cobb angle seem valuable in predicting short-term mortality.