Comparison of two prognostic models for acute pulmonary embolism

Abstract

Pulmonary embolism severity index (PESI) as an example of clinical model which was designed in 2005, which comprises 11 routinely available clinical predictor variables with different prognostic weights. On the basis of PESI score, each patient is classified into one of five classes (I–V), with a 30 day mortality ranging from 1.1% to 24.5%. Patients in risk classes I and II are categorized as low risk and those in risk classes III–V are categorized as high risk. The European society of cardiology (ESC) guidelines have suggested a prognostic model that integrates clinical and laboratory tests which categorize patients, according to individual estimates of pulmonary embolism-related early mortality risk, into high-risk patients (early mortality risk higher than 15%), and non-high-risk patients. However, no study has so far prospectively validated this integrated prognostic model in actual practice. Moreover, the results of most studies supporting the use of biomarkers and echocardiography have not generally been adjusted to established clinical signs of poor outcome. So the aim of this work is to compare the accuracy of PESI and ESC prognostic models in risk stratification of APE patients and their predictive role for short term prognosis, hoping for better risk stratification and therapeutic decision making to improve the outcome of APE. Patients and methods: Forty patients who were confirmed as having APE and met the inclusion criteria were enrolled into this study. They were diagnosed as having acute pulmonary embolism through: (1) thorough medical history, (2) clinical examination, (3) radiological evaluation [plain X-ray and computed tomography pulmonary angiography], (4) arterial blood gases, (5) serum level of D-dimer, (6) routine laboratory investigations and (7) electrocardiogram. All studied patients were subjected to the following: (1) assessment of right ventricular dysfunction by: (a) echocardiography, (b) cardiac biomarkers assessment (serum levels of brain natriuretic peptide and troponin I and 2) Prognostic risk stratification of the patients by PESI and ESC models, then assess the patients outcome. Results: This study showed thirty-day outcome of APE in the form of survival was reported in 90% and adverse events in 20% of the studied patients where mortality (primary outcome) and complications other than mortality (secondary outcome) had an equal rate (10% each). The percentage of patients in low risk groups was nearly the same in PESI and ESC models. The patients in intermediate class in PESI model was lower than those in ESC model. Conversely the percentage of patients in high risk class in PESI was higher than that in ESC model. This study showed that mortality and complications were increased with increased risk classes in both PESI and ESC models. The highest mortality rate was reported in the high risk class which was statistically significant in both models. Mortality prediction of APE patients in this work is statistically highly significant and more accurate among the high risk class in ESC model than in PESI model with 100% specificity and PPV. At cut-off level 100 pg/ml of serum BNP, there were high sensitivity and NPV (100% each) and the highest specificity (91.7%) in predicting the 30-day mortality among APE patients with statistically high significance, in comparison to serum D-dimer and troponin I. Conclusion: (1) There is an agreement to great extent in risk stratification of APE patients by PESI and ESC prognostic models, where mortality rate is increased among high risk classes of both models, (2) ESC prognostic model is more accurate than PESI model in mortality prediction of APE patients especially in the high risk class, (3) echocardiographic evidence of RVD and elevated plasma BNP can help to identify APE patients at increased risk of adverse short-term outcome and (4) integration of RVD assessment by echocardiography and BNP to clinical findings improves the prognostic value of ESC model.

Keywords

• Acute pulmonary embolism
• PESI
• ESC