Abstract
BACKGROUND: Intermediate-risk acute pulmonary embolism (APE) patients are usually defined as hemodynamically stable, comprehending a great therapeutic dilemma. Although anticoagulation therapy is sufficient for most intermediate-risk APE patients, some patients can deteriorate and eventually require a systemic fibrinolytic agent or thrombectomy. Hence, this study aimed to evaluate the predictive value of differences in clinical data for the short-term prognosis of intermediate-risk APE patients. METHODS: A retrospective cohort of 74 intermediate-risk APE patients confirmed by computed tomography pulmonary angiography was analyzed in the present study. Adverse clinical event outcomes included PE-related in-hospital deaths, critical systolic blood pressure consistently under 90 mmHg, refractory to volume loading and vasopressor infusion requirements, mechanical ventilation, and cardiopulmonary resuscitation. The APE patients were stratified into two groups: adverse outcome (n = 25) and control (n = 49) groups. Then, the clinical data of the two groups were compared. Receiver operating characteristic (ROC) curves were used to explore the predictive value of white blood cell (WBC) counts and the right to left ventricular short-axis (RV/LV) ratio. Model calibration was assessed using the Hosmer-Lemeshow goodness-of-fit statistic. RESULTS: The brain natriuretic peptide, WBC count, and the RV/LV ratio were higher in patients with adverse outcomes compared to controls. The APE patients with adverse outcomes presented significantly higher rates of syncope, Negative T waves (NTW) in V1-V3, intermediate-high risk, thrombolytic therapy, and low arterial oxygen saturation (SaO(2)) compared to controls. In the multivariate logistic regression analysis, the SaO(2) < 90%, [odds ratio (OR) 5.343, 95% confidence interval (CI) 1.241-23.008; p = 0.024], RV/LV ratio (OR 7.429, 95% CI 1.145-48.209; p = 0.036), Syncope (OR 12.309, 95% CI 1.702-89.032; p = 0.013), NTW in V1-V3 (OR 5.617, 95% CI 1.228-25.683; p = 0.026), and WBC count (OR 1.212, 95% CI 1.035-1.419; p = 0.017) were independent predictors of in-hospital adverse outcomes among APE patients. The ROC curve analysis indicated that the RV/LV ratio can be used to predict adverse outcomes (AUC = 0.748, p < 0.01) and calibration (Hosmer-Lemeshow goodness of fit test, p = 0.070). Moreover, an RV/LV ratio > 1.165 was predictive of adverse outcomes with sensitivity and specificity of 88.00 and 59.20%, respectively. The WBC counts were also able to predict adverse outcomes (AUC = 0.752, p < 0.01) and calibration (Hosmer-Lemeshow goodness of fit test, p = 0.251). A WBC count > 9.05 was predictive of adverse outcomes with sensitivity and specificity of 68.00 and 73.50%, respectively. CONCLUSION: Overall, a SaO(2) < 90%, RV/LV ratio, Syncope, NTW in V1-V3, and WBC counts could independently predict adverse outcomes in hospitalized intermediate-risk APE patients.