Redo mitral valve replacement for initially rheumatic heart disease: In-hospital outcomes and operative risk prediction

Mohamed Amr; Elsayed  Fayad; Yasser Kamal

doi:10.35810/ects.v4i2.219

Mohamed Amr Department of Cardiothoracic Surgery, Faculty Of Medicine, Suez Canal University, Egypt
Elsayed Fayad Department of Cardiothoracic Surgery, Faculty of Medicine, Suez Canal University, Egypt
Yasser Kamal Department of Cardiothoracic Surgery, Faculty of Medicine, Minia University, El-Minya, Egypt

DOI: https://doi.org/10.35810/ects.v4i2.219

Keywords: Mitral valve surgery, Rheumatic heart disease, Reoperation, Predictors, EuroSCORE II

Abstract

Background: Redo mitral valve replacement (redo-MVR) remains a challenge despite advances in surgical techniques. Little is known about the outcomes of redo-MVR in patients with rheumatic heart disease. We aimed To evaluate the in-hospital outcomes and associated risk factors for mortality and morbidity after re-operative mitral valve replacement in patients with initial rheumatic heart disease.

Methods: This retrospective cohort study included 214 patients, 96 males (44.9%) and 118 females (55.1%), who underwent redo-MVR between January 2015 and December 2020. The mean age was 41.87±11.7 years. European Heart Surgery Risk Assessment System II (EuroSCORE II), Age, Creatinine, Ejection Fraction (ACEF) scores were used for risk stratification. The primary endpoints were in-hospital mortality, major morbidity (renal failure, prolonged ventilation, stroke, reoperation, or deep sternal wound infection), and the composite outcome of mortality and/or morbidity).

Results: Major morbidities occurred in 31.8% of patients, and the in-hospital mortality rate was 19.6%. Predictors of mortality were New York Heart Association class (NYHA) III/IV (OR: 5.4; p˂ 0.001), cardiogenic shock (OR: 13.74, p˂0.001), low left ventricular ejection fraction (LVEF) (OR: 4.36; p= 0.01), and perioperative intra-aortic balloon pump (OR: 6.79; p= 0.01). The significant predictors of mortality and/or major morbidity were NYHA III/IV (OR: 2.39; p˂0.001), low LVEF (OR: 4.44; p= 0.001), active endocarditis (OR: 2.4; p=0.04), and perioperative IABP (OR: 3.88; p= 0.045). EuroSCORE II had better accuracy than the ACEF score to predict adverse outcomes (AUC: 0.70 [95% CI: 0.63-0.78] versus 0.58 [95% CI: 0.50-0.66], p= 0.01) .

Conclusion: Advanced NYHA class and low LVEF could be associated with poor outcomes after redo-MVR in patients with primary surgery for rheumatic mitral valve disease. EuroSCORE II is a helpful tool for risk stratification during redo-MVR.