EUVIMEDEUVIMED entdecken
Systematische ÜbersichtMetaanalyseOpen Access · CC BY 4.0

Machine-Learning Versus Traditional Scores for Predicting Outcomes After Coronary Artery Bypass Graft Surgery: A Systematic Review and Meta-Analysis

Machine Learning versus traditionelle Scores zur Vorhersage von Outcomes nach Koronararterien-Bypass-Operation: systematische Übersicht und Metaanalyse

Aashray K Gupta iD, Ammar Zaka, Daksh Tyagi, Daud Mutahar, Razeen Parvez, Benjamin Muston, Maria Farag, Alexander Lombardo, Aditya Eranki, Ashley Wilson-Smith, Gihwan Song, Brandon Stretton und 9 weitere

Surgical Innovation · 05.08.2026

DOI: 10.1177/15533506261468192 ↗

Auf einen Blick

26 Studien zeigen: ML-Modelle sagten die Mortalität nach CABG besser vorher als traditionelle Risikoscores. Die Diskrimination war höher, die Kalibrierung wurde uneinheitlich berichtet.

Fragestellung

Vergleichen ML-Modelle und traditionelle Risikoscores die Vorhersage von Mortalität nach CABG besser?

Methode

Systematische Übersicht und Metaanalyse; Suche in PubMed, Embase, Web of Science und Cochrane Library bis 14. August 2025; Meta-Analyse mit linear mixed-effects model.

Ergebnisse

26 Studien mit 565 063 Teilnehmer:innen. C-Statistik: 0,82 (95 %-KI 0,79–0,85) für ML vs. 0,73 (95 %-KI 0,71–0,76) für traditionelle Scores; P < 0,0001. Top-ML-Modell: 0,98 (95 %-KI 0,95–1,00).

Einschränkungen

Kalibrierung uneinheitlich berichtet; geringe Zahl gepoolter Studien; I2 = 98 %; überwiegend hohes Verzerrungsrisiko; wenig externe Validierung.

Schlussfolgerung der Autor:innen

ML-Ansätze zeigten eine höhere gepoolte Diskrimination als traditionelle Risikoscores für die Mortalität nach CABG. Die Autor:innen sehen darin einen Hinweis auf Potenzial, nicht den Beleg klinischer Überlegenheit.

Kernaussagen

  • ML-Modelle: C-Statistik 0,82 vs. 0,73 bei traditionellen Scores
  • 26 Studien mit 565 063 Teilnehmer:innen
  • Hohe Heterogenität und wenig externe Validierung

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

Kurz gefasst

Systematische Übersicht in der Zeitschrift Surgical Innovation, online erschienen am 05.08.2026. Verfasst von Aashray K Gupta und 20 weiteren Autor:innen. Beteiligte Einrichtung: Discipline of Surgery, University of Adelaide, Adelaide, South Australia, Australia. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 86 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

Auf einen BlickAbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Coronary artery bypass grafting (CABG) is associated with significant morbidity and mortality. Traditional risk scores, such as the Society of Thoracic Surgery (STS) and EuroSCORE II, have limitations in predicting outcomes, particularly in high-risk patients. Machine learning (ML) models may address these issues by detecting nuanced data patterns not captured by conventional methods. This systematic review and meta-analysis compared the efficacy of ML models with traditional risk scores in predicting outcomes after CABG.

Methods

A comprehensive literature search of records up to August 14 th 2025, was conducted using PubMed, Embase, Web of Science, and the Cochrane Library. Studies included used ML algorithms and traditional risk scores to predict all-cause mortality (in-hospital, 30-day, or longer term as reported by each study) following CABG. Data extraction and quality assessment were independently performed by two reviewers. Meta-analyses were conducted using a linear mixed-effects model, with C-statistics as the primary measure of discrimination.

Results

Twenty-six studies, comprising 565 063 participants, met the inclusion criteria. The pooled C-statistic for ML models was 0.82 (95% CI 0.79-0.85), significantly higher than the 0.73 (95% CI 0.71-0.76) for traditional risk scores ( P < 0.0001). The top-performing ML model achieved a C-statistic of 0.98 (CI 0.95-1.00). Calibration was reported inconsistently across studies and was synthesised narratively rather than quantitatively. Where reported, ML calibration was generally adequate but a robust head-to-head comparison with traditional risk scores was not possible. Subgroup analyses revealed consistent superior performance of ML models across various algorithms and covariate sets.

Conclusions

In this meta-analysis of predominantly internally-validated models, ML approaches showed higher pooled discrimination than traditional risk scores for mortality after CABG. However, given the small number of pooled studies, very high between-study heterogeneity (I 2 = 98%), the predominance of high risk-of-bias studies, and the scarcity of external validation, these findings should be interpreted as supporting the promise of ML rather than establishing proof of clinical superiority. Confirmatory prospective, externally validated studies are required before ML can be recommended for routine pre-operative risk stratification.

Schlagwörter

Stichwörter der Autor:innen

artificial intelligencecoronary artery bypass graftingmachine-learningmeta-analysismortalityrisk prediction

Autor:innen und Institutionen

  • Discipline of Surgery, University of Adelaide, Adelaide, South Australia, Australia.
  • Department of Medicine, Gold Coast University Hospital, Southport, Queensland, Australia.
  • School of Medicine, University of Newcastle, Callaghan, New South Wales, Australia.
  • School of Medicine, Bond University, Varsity Lakes, Queensland, Australia.
  • Department of Cardiothoracic Surgery, Royal Prince Alfred Hospital, New South Wales, Australia.
  • Department of Surgery, Princess Alexandra Hospital, Brisbane, Queensland, Australia.

ORCID-Profile: Aashray K Gupta

Zitieren

APA

Gupta, A. K., Zaka, A., Tyagi, D., Mutahar, D., Parvez, R., Muston, B., Farag, M., Lombardo, A., Eranki, A., Wilson-Smith, A., Song, G., Stretton, B., Kovoor, J. G., Bacchi, S., Ramponi, F., Chan, J. C. Y., Zaman, S., Chow, C., Kovoor, P., …, & Maddern, G. J. (2026). Machine-Learning Versus Traditional Scores for Predicting Outcomes After Coronary Artery Bypass Graft Surgery: A Systematic Review and Meta-Analysis. Surgical Innovation. https://doi.org/10.1177/15533506261468192

Vancouver

Gupta AK, Zaka A, Tyagi D, Mutahar D, Parvez R, Muston B, et al. Machine-Learning Versus Traditional Scores for Predicting Outcomes After Coronary Artery Bypass Graft Surgery: A Systematic Review and Meta-Analysis. Surgical Innovation. 2026. doi:10.1177/15533506261468192

BibTeX
@article{Gupta2026,
  author = {Gupta, Aashray K and Zaka, Ammar and Tyagi, Daksh and Mutahar, Daud and Parvez, Razeen and Muston, Benjamin and Farag, Maria and Lombardo, Alexander and Eranki, Aditya and Wilson-Smith, Ashley and Song, Gihwan and Stretton, Brandon and Kovoor, Joshua G and Bacchi, Stephen and Ramponi, Fabio and Chan, Justin C Y and Zaman, Sarah and Chow, Clara and Kovoor, Pramesh and Bennetts, Jayme S and Maddern, Guy J},
  title = {Machine-Learning Versus Traditional Scores for Predicting Outcomes After Coronary Artery Bypass Graft Surgery: A Systematic Review and Meta-Analysis},
  journal = {Surgical Innovation},
  year = {2026},
  doi = {10.1177/15533506261468192}
}
Kostenlos starten →