To determine predictive factors, clinical and demographics characteristics of patients with pulmonary embolism (PE) in ICU, and to identify factors associated with poor outcome in the hospital and in the ICU.
During a four-year prospective study, a medical committee of six ICU physicians prospectively examined all available data for each patient in order to classify patients according to the level of clinical suspicion of pulmonary thromboembolism. During the study periods, all patients admitted to our ICU were classified into four groups. The first group includes all patients with confirmed PE; the second group includes some patients without clinical manifestations of PE; the third group includes patients with suspected and not confirmed PE and the fourth group includes all patients with only deep vein thromboses (DVTs) without suspicion of PE. The diagnosis of PE was confirmed either by a high-probability ventilation/perfusion (V/Q) scan or by a spiral computed tomography (CT) scan showing one or more filling defects in the pulmonary artery or in its branches. The diagnosis was also confirmed by echocardiography when a thrombus in the pulmonary artery was observed.
During the study periods, 4408 patients were admitted in our ICU. The diagnosis of PE was confirmed in 87 patients (1.9%). The mean delay of development of PE was 7.8 ± 9.5 days. On the day of PE diagnosis, clinical examination showed that 50 patients (57.5%) were hypotensive, 63 (72.4%) have SIRS, 15 (17.2%) have clinical manifestations of DVT and 71 (81.6%) have respiratory distress requiring mechanical ventilation. In our study, intravenous unfractionated heparin was used in 81 cases (93.1%) and low molecular weight heparins were used in 4 cases (4.6%). The mean ICU stay was 20.2 ± 25.3 days and the mean hospital stay was 25.5 ± 25 days. The mortality rate in ICU was 47.1% and the in-hospital mortality rate was 52.9%. Multivariate analysis showed that factors associated with a poor prognosis in ICU are the use of norepinephrine and epinephrine. Furthermore, factors associated with in-hospital poor outcome in multivariate analysis were a number of organ failure associated with PE ≥ 3.
Moreover, comparison between patients with and without pe showed that predictive factors of pe are: acute medical illness, the presence of meningeal hemorrhage, the presence of spine fracture, hypoxemia with PaO2/FiO2 ratio <300 and the absence of pharmacological prevention of venous thromboembolism.
Despite the high frequency of DVT in critically ill patients, symptomatic PE remains not frequently observed, because systematic screening is not performed. Pulmonary embolism is associated with a high ICU and in-hospital mortality rate. Predictive factors of PE are acute medical illness, the presence of meningeal hemorrhage, the presence of spine fracture, hypoxemia with PaO2/FiO2 < 300 and the absence of pharmacological prevention of venous thromboembolism.