Miguel A. Quintana et al. Joaquín Perea Ricardo Villarreal Mauricio Soto … Marildes Castro Rafael Cortês Patricio Sanhueza Iván Saavedra Yhaquelina Quispe Roger Tenezaca Roberto Colque Alexandra Arias Alex Rivera Toquica Manuel Álvarez Ysmenia D. Pérez Paola Morejón Karina G. Fernández Adriana Mogollón Marcelo Cardona Gonzalo Emanuel Pérez Arnaldo López-Ruiz
Abstract
Background
Risk stratification in acute pulmonary embolism (PE) is based primarily on findings obtained at hospital admission. However, whether serial assessment of right ventricular (RV) dysfunction and hemodynamic changes provides incremental prognostic value beyond the initial risk stratification remains uncertain.
Objectives
To evaluate longitudinal clinical and imaging variables associated with adverse in-hospital outcomes in patients with intermediate-high- and high-risk acute PE.
Methods
This prospective registry included patients with intermediate-high- and high-risk acute PE. The primary composite endpoint comprised in-hospital mortality, invasive mechanical ventilation, persistent shock or hypotension, and extracorporeal membrane oxygenation. Longitudinal variables were assessed 24-72 hours after admission. All statistical tests were two-tailed, and a p-value < 0.05 was considered statistically significant. Multivariable logistic regression models were constructed, and discriminative performance was evaluated using the area under the receiver operating characteristic curve (AUC-ROC). A prespecified missing-data analysis was performed to assess the availability of follow-up computed tomography pulmonary angiography (CTPA).
Results
The primary composite endpoint occurred in 21.2% of the 151 patients, with an in-hospital mortality rate of 9.3%. Persistent RV dysfunction on follow-up CTPA (χ2 = 15.48; p < 0.001) and follow-up systolic blood pressure (SBP; p < 0.001) were the strongest individual predictors of adverse outcomes. In the longitudinal multivariable model (n = 60), persistent RV dysfunction on follow-up CTPA (OR 9.70; 95%CI, 1.96-47.89; p = 0.005) and follow-up SBP (OR 0.90; 95%CI, 0.83-0.96; p = 0.003) remained independent predictors. The model demonstrated excellent discrimination (AUC-ROC = 0.874) and adequate calibration. The missing-data analysis showed that patients without follow-up CTPA were more severely ill at baseline, consistent with a missing not at random mechanism.
Conclusions
Reassessment of established risk markers during hospitalization may improve identification of patients who remain at high risk despite initial risk classification and treatment. Persistent RV dysfunction and lower follow-up SBP were independently associated with adverse in-hospital outcomes, which supports the use of dynamic risk stratification in acute PE.
