Machine Learning Analysis Reveals Clinical Features and Risk Factors Associated with Adverse Clinical Outcomes in Central PE

Jun 9, 2023

The following is a summary of "Clinical features and risk factors of adverse clinical outcomes in central pulmonary embolism using machine learning analysis," published in the May 2023 issue of Pulmonology by Martinez, et al.

An independent predictor for thrombolysis in earlier research, central pulmonary embolism (PE), was connected to high clot burden. More knowledge about the indicators of unfavorable outcomes in these patients was required for better risk classification. For a study, researchers sought to identify independent predictors of adverse clinical outcomes in patients with central PE, to improve risk stratification and management.

It was a large retrospective, observational, and single-center study that included hospitalized patients with central PE. Data on patient demographics, comorbidities, clinical features on admission, imaging results, treatments, and outcomes were collected. Multivariable standard logistic regressions and the Least Absolute Shrinkage and Selection Operator (LASSO) machine learning logistic regressions were utilized to analyze the factors associated with a composite of adverse clinical outcomes, including vasopressor use, mechanical ventilation, and inpatient mortality. Sensitivity analyses were also performed.

A total of 654 patients diagnosed with central PE were included in the study. The mean age of the patients was 63.1 years, with 59% being women and 82% African American. The composite adverse outcome occurred in 18% (n = 115) of the patients. Several independent predictors of adverse clinical outcomes were identified, including elevation of serum creatinine (odds ratio [OR] = 1.37, 95% CI = 1.20–1.57; P = 0.0001), the elevation of white blood cell (WBC) count (OR = 1.10, 95% CI = 1.05–1.15; P < 0.001), higher simplified pulmonary embolism severity index (sPESI) score (OR = 1.47, 95% CI = 1.18–1.84; P = 0.001), elevation of serum troponin (OR = 1.26, 95% CI = 1.02–1.56; P = 0.03), and increase in respiratory rate (OR = 1.03, 95% CI = 1.0–1.05; P = 0.02). However, right ventricular dysfunction on imaging and saddle PE location did not predict adverse outcomes.

Several independent predictors of adverse clinical outcomes were identified among patients with central PE, including elevated serum creatinine, elevated WBC count, higher sPESI score, elevated serum troponin, and increased respiratory rate. The findings provided valuable insights for risk stratification and clinical management of patients with central PE, highlighting the importance of early recognition and appropriate intervention.

Source: resmedjournal.com/article/S0954-6111(23)00183-X/fulltext