Abstract
Background
Decreased ejection fraction (EF) has been found to be a strong predictor of mortality. However, this association has not been studied in patients undergoing ventriculography over a long period of time. Using a large database of patients who underwent coronary angiography for clinical reasons, we evaluated any association between severity of EF impairment and 10 year mortality.
Methods
Retrospective angiographic data of 1,937 patients between 1993 to 1997 from the VA Long Beach Health Care System were studied. Decreased EF was defined as EF<50%. Using chi-square test and Cox survival analysis, we evaluated any association between decreased EF with mortality. Furthermore, we evaluated different degrees of decreased EF with all-cause mortality using uni- and multivariate analysis.
Results
Total mortality was 22.9% of the cohort. Decreased EF was a strong predictor of death over a period of 10 years. (All-cause mortality occurred in 21.6 % of patients with normal EF vs. 41.7 %, OR 2.59, CI 2.06-3.26, P <.001). After adjustment for left main coronary artery disease, mitral regurgitation, three-vessel coronary disease, and clinical risk factors (diabetes, hypertension, hyperlipidemia, and smoking), decreased EF remained independently associated with all cause-mortality regardless of ethnicity.
Conclusion
Decreased EF measured during coronary angiography is a strong independent predictor of all-cause mortality over a period of 10 years.
1
Background
An estimated 3–4 million people have coronary artery disease with decreased ejection fraction (EF). This number is increasing exponentially with 400,000 new cases each year . Prior studies have addressed prognostic significance of decreased EF after myocardial infarction with and without intervention (thrombolytics and percutaneous coronary intervention) and in congestive heart failure . Limiting factors in prior literature were small number of patients in the studies and short follow-up periods. Our goal was to evaluate prognostic effect of low EF on mortality in a large group of patients over a long period. Furthermore, we analyzed different degrees of EF association with mortality and performed multivariate analysis adjusting for baseline risk factors and severity of coronary artery disease.
2
Methods
Using a large database of patients who underwent coronary angiography and ventriculography for clinical reasons between years 1993 and 1997 at the VA Long Beach Health Care System, we evaluated any association between decreased EF and the severity of low EF with 10-year mortality. The protocol was approved by the institutional review board. Follow-up data was available over a period of 10 years.
Left ventricular (LV) EF was calculated using ventriculography performed at the time of coronary angiography. Decreased EF was defined as LVEF <50%.
The chi-square test was used to compare nominally scaled variables. Survival data were analyzed using the Cox survival analysis. Logistic regression analysis was utilized for multivariate analysis. A P value of .05 or less was deemed as significant.
2
Methods
Using a large database of patients who underwent coronary angiography and ventriculography for clinical reasons between years 1993 and 1997 at the VA Long Beach Health Care System, we evaluated any association between decreased EF and the severity of low EF with 10-year mortality. The protocol was approved by the institutional review board. Follow-up data was available over a period of 10 years.
Left ventricular (LV) EF was calculated using ventriculography performed at the time of coronary angiography. Decreased EF was defined as LVEF <50%.
The chi-square test was used to compare nominally scaled variables. Survival data were analyzed using the Cox survival analysis. Logistic regression analysis was utilized for multivariate analysis. A P value of .05 or less was deemed as significant.
3
Results
One thousand nine hundred thirty-seven patients who underwent coronary angiography and ventriculography between 1993 and 1997 were included in the study. Race and age distribution of the study population are shown in Figs. 1 and 2 . Total mortality was 22.9% in the study population. Decreased EF was a strong predictor of death over a period of 10 years. (All-cause mortality occurred in 21.6% of patients with normal EF vs. 41.7%, OR 2.59, CI 2.06–3.26, P <.001). After adjustment for age, left main coronary artery disease, mitral regurgitation, three-vessel coronary disease, and clinical risk factors (diabetes mellitus, hypertension, hyperlipidemia, and smoking) and age, decreased EF remained independently associated with all-cause mortality. Separating decreased EF to mild (EF 45–54%), moderate (EF 35–44%), or severe (EF <35%), mild to moderate decreased in EF also showed independent association with lower survival but to a lesser degree in comparison to severely decreased EF ( Figs. 3 and 4 ). This trend was similar between Caucasians and other ethnicities ( Figs. 5 and 6 ).
