Predictors of Coronary Artery Disease in Heart Failure with Reduced Ejection Fraction at the Aga Khan University Hospital in Nairobi

Redemptar Kimeu, Mohamed Jeilan, Mzee Ngunga

Research output: Contribution to journalArticlepeer-review


There appears to be an epidemiological transition in the etiology of heart failure in sub- Saharan Africa (SSA) in parallel with a steady increase in risk factors for coronary artery disease (CAD). SSA has limited access to heart failure and CAD diagnostics, limiting the number of patients who receive optimal care. Our objectives were to study the predictors of coronary artery disease among patients with heart failure with reduced ejection fraction (HFrEF) and develop a model to assist clinicians in determining the likelihood of CAD before cardiac catheterization. Methodology: This was a retrospective study at the Aga Khan University Hospital, Nairobi, which is equipped with diagnostic capabilities for heart failure and coronary artery assessment. We evaluated patients with HFrEF based on echocardiographic data over a 12-year period. Patients with coronary anatomical evaluation data were included. A multivariable model of CAD was generated using stepwise logistic regression. Results: Of the 1329 patients screened, 514 met the inclusion criteria. The mean age was 61.0 ± 12.8 years. There were 381 male cases (75.2%), and the predominant race was African, numbering 386 (75.2%). Most patients, 97%, were evaluated through conventional coronary angiography. Further, 310 (60.3%) cases had significant CAD. The prevalence of CAD in HFrEF was 52.3% in Africans, 85% in Asians, and 79% in Caucasians. In the multivariable logistic regression, the odds of having significant CAD was higher among participants with diabetes mellitus (aOR: 1.86; 95%CI: 1.15-3.03), Q waves (aOR: 2.12; 95%CI: 1.12-4.10), significant ST segment deviation (aOR: 4.14; 95%CI: 2.23-8.03), and regional wall motion abnormalities on echocardiogram (aOR: 6.53; 95%CI: 3.94-11.06). Conclusion: In this population, CAD was a major etiology in HFrEF among the African population. The most powerful predictors of CAD were type 2 diabetes, the presence of pathological Q waves, or ST segment shift on a 12-lead electrocardiogram, and regional wall motion abnormality on 2D echocardiogram.

Original languageEnglish
Article number56
JournalGlobal Heart
Issue number1
Publication statusPublished - 2023


  • HFrEF
  • coronary artery disease
  • predictors


Dive into the research topics of 'Predictors of Coronary Artery Disease in Heart Failure with Reduced Ejection Fraction at the Aga Khan University Hospital in Nairobi'. Together they form a unique fingerprint.

Cite this