Risk stratification in aortic arch surgery remains a significant challenge because of the heterogeneity of clinical presentations and surgical approaches. Existing models such as Society of Thoracic Surgeons and European System for Cardiac Operative Risk Evaluation II often lack precision in this subset. The ARCH score was developed to predict morbidity and mortality in patients undergoing aortic arch surgery with hypothermic circulatory arrest, regardless of urgency or procedural complexity. However, its external validation has been limited. Methods: This study represents the first multicenter external validation of the ARCH score in a mixed European and North American population. A total of 851 consecutive patients who underwent aortic arch surgery of any urgency with hypothermic circulatory arrest in 2 centers between 2013 and 2024 were retrospectively analyzed. Discriminatory ability and calibration were assessed using receiver operating characteristic curve, bootstrap validation, and Brier scores. Comparative analysis of the populations was also conducted against the original ARCH score cohort and a previous single-center validation. Results: The ARCH score demonstrated excellent discrimination (c-statistic 0.798) and good calibration (Brier score 0.09) for mortality. Society of Thoracic Surgeons morbidity and mortality predictions showed fair discrimination (c-statistic 0.724, Brier score 0.12). Comparative analysis showed consistency with the original study and suggested that lower-than-expected mortality may contribute to the reduced discriminatory power observed in the previous validation cohort, despite high predictive accuracy. Conclusion: The ARCH score is a reliable and broadly applicable risk prediction tool in aortic arch surgery, including high-risk and emergent settings. Its simplicity and strong performance support wider adoption, particularly with the future development of an accessible online calculator.
Dual-center external validation of the ARCH score: Predictive accuracy and calibration for hypothermic circulatory arrest aortic arch surgery / Cabrucci, Francesco; Baudo, Massimo; Sicouri, Serge; Bacchi, Beatrice; Magouliotis, Dimitrios E.; Yamashita, Yoshiyuki; Chiarello, Bruno; Petrone, Dario; Bessi, Giulia; Pacini, Tommaso; Rodriguez, Roberto; Torregrossa, Gianluca; Bonacchi, Massimo; Gelsomino, Sandro; Ramlawi, Basel. - In: SURGERY. - ISSN 0039-6060. - STAMPA. - (2025), pp. 109971-109978. [10.1016/j.surg.2025.109971]
Dual-center external validation of the ARCH score: Predictive accuracy and calibration for hypothermic circulatory arrest aortic arch surgery
Cabrucci, Francesco;Bacchi, Beatrice;Chiarello, Bruno;Petrone, Dario;Bessi, Giulia;Pacini, Tommaso;Bonacchi, Massimo
;Gelsomino, Sandro;
2025
Abstract
Risk stratification in aortic arch surgery remains a significant challenge because of the heterogeneity of clinical presentations and surgical approaches. Existing models such as Society of Thoracic Surgeons and European System for Cardiac Operative Risk Evaluation II often lack precision in this subset. The ARCH score was developed to predict morbidity and mortality in patients undergoing aortic arch surgery with hypothermic circulatory arrest, regardless of urgency or procedural complexity. However, its external validation has been limited. Methods: This study represents the first multicenter external validation of the ARCH score in a mixed European and North American population. A total of 851 consecutive patients who underwent aortic arch surgery of any urgency with hypothermic circulatory arrest in 2 centers between 2013 and 2024 were retrospectively analyzed. Discriminatory ability and calibration were assessed using receiver operating characteristic curve, bootstrap validation, and Brier scores. Comparative analysis of the populations was also conducted against the original ARCH score cohort and a previous single-center validation. Results: The ARCH score demonstrated excellent discrimination (c-statistic 0.798) and good calibration (Brier score 0.09) for mortality. Society of Thoracic Surgeons morbidity and mortality predictions showed fair discrimination (c-statistic 0.724, Brier score 0.12). Comparative analysis showed consistency with the original study and suggested that lower-than-expected mortality may contribute to the reduced discriminatory power observed in the previous validation cohort, despite high predictive accuracy. Conclusion: The ARCH score is a reliable and broadly applicable risk prediction tool in aortic arch surgery, including high-risk and emergent settings. Its simplicity and strong performance support wider adoption, particularly with the future development of an accessible online calculator.| File | Dimensione | Formato | |
|---|---|---|---|
|
1-s2.0-S0039606025008232-main.pdf
Accesso chiuso
Descrizione: Full text
Tipologia:
Pdf editoriale (Version of record)
Licenza:
Tutti i diritti riservati
Dimensione
1.94 MB
Formato
Adobe PDF
|
1.94 MB | Adobe PDF | Richiedi una copia |
I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.



