Combination of D-dimer and simplified pulmonary embolism severity index to improve prediction of hospital death in patients with acute pulmonary embolism

Predicting acute and long-term mortality in a cohort of pulmonary embolism patients using machine learning

BACKGROUND

Pulmonary embolism (PE) is a severe condition that causes significant mortality and morbidity. Due to its acute nature, scores have been developed to stratify patients at high risk of 30-day mortality. Here we develop a machine-learning based score to predict 30-day, 90-day, and 365-day mortality in PE patients.

METHODS

The Birmingham and Black Country Venous Thromboembolism registry (BBC-VTE) of 2183 venous thromboembolism patients is used. Random forests were trained on a 70% training cohort and tested against 30% held-out set. The outcomes of interest were 30-day, 90-day, and 365-day mortality. These were compared to the pulmonary embolism severity index (PESI) and simplified pulmonary embolism severity index (sPESI). Shapley values were used to determine important predictors. Oral anticoagulation at discharge was also investigated as a predictor of mortality.

RESULTS

The machine learning risk score predicted 30-day mortality with AUC 0.71 [95% CI: 0.63 - 0.78] compared to the sPESI AUC of 0.65 [95% CI: 0.57 - 0.73] and PESI AUC of 0.64 [95% CI: 0.56 - 0.72]. 90-day mortality and 365-day mortality were predicted with an AUC of 0.74 and 0.73 respectively. High counts of neutrophils, white blood cell counts, and c-reactive protein and low counts of haemoglobin were important for 30-day mortality prediction but progressively lost importance with time. Older age was an important predictor of high risk throughout.

CONCLUSION

Machine learning algorithms have improved on standard clinical risk stratification for PE patients. External cohort validation is required before incorporation into clinical workflows.

A New Index for the Prediction of 30-Day Mortality in Patients With Pulmonary Embolism: The Pulmonary Embolism Mortality Score (PEMS)

Our aim was to analyze possibility of combination of basic clinical and radiological signs to predict 30-day mortality after acute pulmonary embolism (PE). We included 486 patients. Age, gender, simplified pulmonary embolism index (sPESI), pH, troponin, N-terminal natriuretic peptide, minimal systolic and diastolic blood pressure, O2 saturation, syncope, need for vasopressors, thrombotic obstruction, vessel diameter, short axis ratio right ventricle/left ventricle, and contrast medium reflux into the inferior vena cava (IVC) were analyzed. A backward algorithm in a logistic regression model was used to identify relevant risk factors. Multiple logistic regression analysis identified that sPESI, pH, minimal diastolic blood pressure, IVC reflux, and need for vasopressors influenced 30-day mortality. A score for mortality prediction was constructed (the Pulmonary Embolism Mortality Score): sPESI >2 points (1 point), pH <7.35 (1 point), minimal diastolic blood pressure <45 mm Hg (1 point), IVC reflux (1 point), and need for vasopressors (2 points). Patients with >3 points showed higher 30-day mortality (sensitivity: 84.9%, specificity: 83.0%, positive predictive value: 51.8%, negative predictive value: 96.2%). The net reclassification improvement compared with the sPESI was 0.94 (95% CI = 0.73-1.15). In conclusion, a new score can predict 30-day mortality in patients with PE and is more sensitive than sPESI.

Trends in risk stratification, in-hospital management and mortality of patients with acute pulmonary embolism: an analysis from China pUlmonary thromboembolism REgistry Study (CURES)

BACKGROUND

Similar trends of management and in-hospital mortality of acute pulmonary embolism (PE) have been reported in European and American populations. However, these tendencies were not clear in Asian countries.

OBJECTIVES

We retrospectively analyzed the trends of risk stratification, management and in-hospital mortality for patients with acute PE through a multicenter registry in China (CURES).

METHODS

Adult patients with acute symptomatic PE were included between 2009 and 2015. Trends in disease diagnosis, treatment and death in hospital were fully analyzed. Risk stratification was retrospectively classified by hemodynamical status and the simplified Pulmonary Embolism Severity Index (sPESI) score according to the 2014 European Society of Cardiology/European Respiratory Society guidelines.

RESULTS

Among overall 7438 patients, the proportions with high (hemodynamically instability), intermediate (sPESI≥1) and low (sPESI=0) risk were 4.2%, 67.1% and 28.7%, respectively. Computed tomographic pulmonary angiography was the widely employed diagnostic approach (87.6%) and anticoagulation was the frequently adopted initial therapy (83.7%). Between 2009 and 2015, a significant decline was observed for all-cause mortality (from 3.1% to 1.3%, adjusted P _{for trend}=0.0003), with a concomitant reduction in use of initial systemic thrombolysis (from 14.8% to 5.0%, P _{for trend}<0.0001). The common predictors for all-cause mortality shared by hemodynamically stable and unstable patients were co-existing cancer, older age, and impaired renal function.

CONCLUSIONS

The considerable reduction of mortality over years was accompanied by changes of initial treatment. These findings highlight the importance of risk stratification-guided management throughout the nation.