Pulmonary embolism (PE) remains one of the leading causes of mortality among cardiovascular
diseases. We aimed at investigating risk factors of PE complications in patients with
intermediate risk and integrate them into a simple model for its' bedside prediction.
Among 173 patients with PE, 136 were classified as high or intermediate risk. Patients
were retrospectively divided into groups of complicated (n = 44) or uncomplicated
(n = 92) course. Study endpoints: obstructive shock, recurrent PE, needs for resuscitation/thrombolysis/hemodynamic
support and death during 30 days.
Predictors of PE complications were: chronic heart failure, diabetes mellitus (DM),
atrial fibrillation, permanent risk factor of venous thromboembolism, syncope, positive
heart-type fatty acid binding protein (hFABP), positive troponin I, heart rate (HR) ≥ 110 bpm,
systolic blood pressure (SBP) ≤ 100 mmHg, creatinine clearance ≤ 70 ml/min. Multivariate
logistic regression analysis was used to model a simple predictive score named ROCky
(Risk of Complications): HR ≥ 110 bpm (1.5 points), SBP ≤ 100 mmHg (2.5 points), positive
hFABP (2 points) and presence of DM (2.5 points). The AUROC of this model was 0.89
to predict any complication, 0.83 for obstructive shock and 0.92 for death from any
cause; the optimal cut-off scores for any complication was ≥2.5 points, ≥3.5 for obstructive
shock and ≥4.5 points for death within 30 days.
hFABP, tachycardia, hypotension and DM were identified as the major independent determinants
of complications development in patients with pulmonary embolism and may be used in
combination as the bedside simple predictive ROCky score for early risk stratification
in intermediate-risk group.