The TEP-HUC Score accurately predicted in-hospital mortality in patients with pulmonary embolism, outperforming the sPESI score (AUC 0.83 [95% CI 0.76-0.89] vs 0.78).
Observational
No
Does the TEP-HUC Score improve the prediction of in-hospital mortality compared to sPESI in patients with pulmonary embolism?
The novel TEP-HUC score, which incorporates clinical and biochemical variables, accurately predicts in-hospital mortality in intermediate-high risk pulmonary embolism and outperforms the standard sPESI score.
Estimación del efecto: AUC 0.83 (95% CI 0.76-0.89)
Abstract Pulmonary embolism (PE) is a frequent condition that requires accurate risk stratification to guide optimal therapeutic strategies. Current prognostic models may underestimate risk in patients with haemodynamic compromise or submassive right ventricular dysfunction. A clinical and analytical score (TEP-HUC) was developed based on variables identified in multivariable analysis. Methods: Logistic regression analysis was performed using in-hospital mortality as the outcome variable. Included predictors were: simplified PESI, lactate 2 mmol/L, SaO₂ 90% (eventhough it is already included in sPESI), vasopressor use, ventilatory support (non-invasive or invasive), renal failure, and NT-proBNP 3, 000 pg/mL. β coefficients were converted into proportional integer points and the following logistic equation was derived: logit (p) = −4. 22 + 0. 82·PESI + 1. 10·lactate + 0. 95·SaO₂ + 2. 35·vasopressors + 2. 10·ventilation + 1. 30·renal failure + 0. 90·NT-proBNP p = 1 / (1 + e^ (−logit (p) ) ) The individual probability and total score (range 0–15 points) were implemented in an interactive Excel spreadsheet for clinical use. Results: The model showed good overall performance (AUC = 0. 83 95% CI 0. 76–0. 89), with adequate calibration (Hosmer–Lemeshow χ² = 6. 2, p = 0. 62) and a Brier score of 0. 082. Point-based stratification defined four risk categories: low (≤4 points, 1. 8%), intermediate (5–7 points, 6. 4%), high (8–10 points, 18%), and very high (≥11 points, 42%). Internal calibration was satisfactory, and discrimination outperformed the sPESI (AUC 0. 83 vs 0. 78). The optimal cut-off (p ≥ 0. 18) achieved a sensitivity of 85% and specificity of 72%. Conclusions: The TEP-HUC Score, which incorporates easily obtainable clinical and biochemical variables, accurately predicts in-hospital mortality in patients with intermediate-high risk PE. These results exceed the ones obtained with sPESI and can be easily implemented in daily clinical practice through an automated spreadsheet tool developed at our centre. Future studies may be needed to validate this score in the clinical practice in order to determine whether it represents a better risk stratification strategy in intermediate-high risk patients. TEP-HUC SCORE. Variables and points ROC curves
Olivero et al. (Fri,) conducted a observational in Pulmonary embolism. TEP-HUC Score vs. sPESI was evaluated on In-hospital mortality (AUC 0.83, 95% CI 0.76-0.89). The TEP-HUC Score accurately predicted in-hospital mortality in patients with pulmonary embolism, outperforming the sPESI score (AUC 0.83 [95% CI 0.76-0.89] vs 0.78).