麻豆传媒

Higher is not necessarily better when it comes to left ventricular ejection fraction (LVEF), according to a retrospective study.

Compared with an LVEF of 60-65%, both reduced and elevated LVEFs were associated with worse all-cause mortality that persisted after multivariable adjustment, Brandon Fornwalt, MD, PhD, of Geisinger Health in Danville, Pennsylvania, and colleagues reported in .

Their findings, over a median of 4 years follow-up, revealed a U-shaped relationship with mortality:

• LVEF 35-40%: adjusted HR 1.73 (95% CI 1.66-1.80)
• LVEF 55-60%: adjusted HR 1.06 (95% CI 1.04-1.08)
• LVEF 65-70%: adjusted HR 1.17 (95% CI 1.14-1.20)
• LVEF ≥70%: adjusted HR 1.71 (95% CI 1.64-1.77)

Deviation was "associated with poorer survival regardless of age, sex, or other relevant comorbidities such as heart failure," they concluded.

Based on their findings -- which examined Geisinger patients from 1998 to 2018 who got echocardiograms most commonly for chest pain, dyspnea or fatigue, coronary artery disease, aortic valve disease, and congestive heart failure -- the group proposed a fourth LV phenotype, heart failure with supra-normal LVEF.

Importantly, patients with LVEF 70% or higher shared a similar mortality risk with the LVEF 35-40% group, which Fornwalt's team said could not be entirely accounted for by differences in heart rate, blood pressure, left ventricular volume, wall thickness, mitral regurgitation, anemia, or hyperthyroidism.

Thus, very-high LVEFs may not be so much as they are associated with increased mortality, according to the authors.

In the subgroup of patients with heart failure, the U-shaped relationship still existed, such that the lowest risk was observed at LVEF 60-65% for outpatients and 55-60% for inpatients.

"LVEF ≥70% predicted a higher mortality amongst both inpatients and outpatients with heart failure, as well as in the larger population without a diagnosis of heart failure, even after adjusting for many confounders, suggesting that these observations are unlikely to be spurious," Fornwalt and colleagues said. "Moreover, heart failure may be underdiagnosed in patients with a supra-normal LVEF as this is not currently a clinically recognized entity."

Geisinger patients in the study totaled more than 203,000 and underwent 403,977 echocardiograms as a group. Participants averaged 64 years at enrollment and 52% were men.

"We believe this is the first analysis exploring the relationship between mortality and routinely reported echocardiographic LVEF, the result that clinicians rely on most to make decisions," Fornwalt's group maintained.

Investigators reproduced the study's main findings using a group of nearly 36,000 people from New Zealand, though this validation cohort was missing too much information for statistical adjustment.

Additional limitations of the study included the likelihood of LVEF measurement errors in the Geisinger dataset, which failed to capture advanced techniques such as 3D echocardiography. Another was that 98% of participants were of European ancestry, limiting the generalizability of the results.

Comment