Cox versus log-logistic models for long-term survival in patients undergoing coronary artery bypass grafting with and without coronary endarterectomy
Folia Medica · 2026
DOI: 10.3897/folmed.68.e168772 ↗Kurz gefasst
Fachartikel in der Zeitschrift Folia Medica, 2026. Verfasst von Feridoun Sabzi und 3 weiteren Autor:innen. Beteiligte Einrichtung: Cardiovascular Research Center, Kermanshah University of Medical Sciences, Kermanshah, Iran. Hauptthemen laut MEDLINE: Endarterectomy, Coronary Artery Bypass, Off-Pump, Coronary Artery Bypass. Die Arbeit nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Introduction
Survival outcomes after coronary endarterectomy (CE) in off-pump coronary artery bypass grafting (OPCAB) remain controversial due to variable complications. Selecting optimal survival models is challenging, especially in underrepresented Middle Eastern cohorts with high CE rates.
Aim
This study compares Cox and log-logistic regression models for predicting long-term survival in OPCAB patients with and without CE, providing insights into risk stratification for complex coronary artery disease (CAD). Materials and methods : A retrospective cohort study analyzed 252 OPCAB patients (46 with CE) at Imam Ali Hospital, Iran, from 2018 to 2019. Data were retrieved from medical records, with follow-up (mean 68.93±19.33 months) via telephone interviews. Variables, defined per EuroSCORE 2, included demographics, comorbidities, and postoperative outcomes. Cox and log-logistic models were evaluated using Cox-Snell residual plots, Akaike Information Criterion (AIC), and Receiver Operating Characteristic (ROC) curves to assess model performance and discrimination, with analyses in STATA and SPSS ( p <0.05).
Results
Log-logistic regression identified age, gender, angina, preoperative antiplatelet therapy, obtuse marginal 3 disease, atrial fibrillation, and ejection fraction as significant mortality predictors. CE patients had longer intensive care unit (ICU) stays, higher transfusion needs, and increased arrhythmias versus OPCAB-only patients. Cox-Snell plots, lower AIC (1308 vs. 1330 for Cox), and superior ROC curve performance confirmed improved log-logistic model fit.
Conclusion
This study provides early evidence in a Middle Eastern cohort that the log-logistic model offers a better fit than the Cox model for predicting survival in OPCAB patients with high CE rates. These findings advocate parametric models for survival analysis in diffuse CAD, enhancing risk stratification in resource-limited settings like Iran.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Cardiovascular Research Center, Kermanshah University of Medical Sciences, Kermanshah, Iran.
- Department of Biostatistics, School of Health, Kermanshah University of Medical Sciences, Kermanshah, Iran.
- Students Research Committee, School of Health, Kermanshah University of Medical Scinces, Kermanshah, Iran.
ORCID-Profile: Atefeh Asadmobini
Zitieren
APA
Sabzi, F., Asadmobini, A., Mahaki, B., & Solouki, L. (2026). Cox versus log-logistic models for long-term survival in patients undergoing coronary artery bypass grafting with and without coronary endarterectomy. Folia Medica. https://doi.org/10.3897/folmed.68.e168772
Vancouver
Sabzi F, Asadmobini A, Mahaki B, Solouki L. Cox versus log-logistic models for long-term survival in patients undergoing coronary artery bypass grafting with and without coronary endarterectomy. Folia Medica. 2026. doi:10.3897/folmed.68.e168772
BibTeX
@article{Sabzi2026,
author = {Sabzi, Feridoun and Asadmobini, Atefeh and Mahaki, Behzad and Solouki, Leila},
title = {Cox versus log-logistic models for long-term survival in patients undergoing coronary artery bypass grafting with and without coronary endarterectomy},
journal = {Folia Medica},
year = {2026},
doi = {10.3897/folmed.68.e168772}
}
EUVIMED