A systematic review and meta-analysis of population-based echocardiographic studies has established updated reference ranges for left ventricular ejection fraction in healthy adults, identifying lower limits of normality of approximately 53 percent for women and 52 percent for men. Conducted by researchers including Anne Emilie Morsing, Filip Gnesin, Asya Lyass, and Charlotte Andersson, the investigation pooled data across ten community-based cohorts encompassing 10,427 healthy participants. Across the entire pooled population, the overall mean left ventricular ejection fraction was 62.8 percent, with a 95 percent confidence interval spanning 61.0 percent to 64.7 percent, yielding estimated central 95 percent reference limits ranging from a lower bound of 51.8 percent or 52.2 percent to an upper bound of 73.2 percent or 73.5 percent. A central conclusion from the statistical distribution is that an ejection fraction below 50 percent is highly unlikely to represent normal cardiac function, carrying a probability of less than 5 percent of falling within the normal physiological range across every demographic group evaluated.
Transthoracic echocardiography-derived left ventricular ejection fraction serves as a foundational measurement in clinical cardiology and heart failure risk prevention, reflecting the proportion of blood pumped out of the left ventricle with each contraction. In conditions such as dilated cardiomyopathy and heart failure, subtle myocardial changes develop years before individuals experience clinical symptoms, presenting a vital window of opportunity for early, disease-modifying interventions. Determining which individuals should undergo targeted clinical management requires precisely defined boundaries between normal and abnormal systolic function. The widely referenced 2015 recommendations jointly issued by the American Society of Echocardiography and the European Association of Cardiovascular Imaging designated thresholds below 52 percent in men and below 54 percent in women as suggestive of abnormal function, but those recommendations drew predominantly from select cohorts of White individuals. Defining accurate reference values remains critical to guiding preventive cardiovascular strategies while preventing unnecessary downstream diagnostic investigations and treatments among people who are healthy.
Subsequent investigations, such as the World Alliance Society of Echocardiography Normal Values Study initiative, indicated that normal ejection fractions might vary across racial groups and could be substantially higher than guideline cutoffs, suggesting normative thresholds of 58 percent in women and 57 percent in men. However, when evaluated in a large real-world clinical cohort undergoing transthoracic echocardiography, long-term mortality rates appeared identical between patients classified as normal exclusively by American Society of Echocardiography and European Association of Cardiovascular Imaging criteria and those who met both guideline and World Alliance Society of Echocardiography criteria. This observation raised concerns that higher thresholds might inappropriately classify healthy people as abnormal. To clarify these discrepancies, the authors sought to derive demographically sensitive reference intervals across sex, age, and self-reported race and ethnicity by restricting their synthesis to healthy, population-based adult cohorts.
The study protocol followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Two independent reviewers systematically searched the PubMed and MEDLINE databases on January 3, 2025, screening records published between January 1, 2000, and the search date using predefined search strings. Eligible investigations were required to be English-language original articles reporting sex-specific mean and standard deviation values of left ventricular ejection fraction measured by two-dimensional or three-dimensional transthoracic echocardiography in healthy adults aged 18 years or older. To prevent biased estimates, the authors excluded pediatric cohorts, case-control control groups, clinical cohorts referred for echocardiography due to suspected cardiac disease even when overt disease was ruled out, and studies employing non-ultrasound imaging modalities such as cardiac magnetic resonance imaging or single-photon emission computed tomography. Out of 1,776 screened records, ten studies met the complete eligibility criteria.
Data extraction captured participant sample sizes, proportion of female participants, mean age, country, ultrasound equipment, and imaging methods. Across the ten cohorts, female representation ranged from 36 percent to 59 percent, representing 4,974 women and 5,453 men, while mean cohort ages ranged from 27 to 54 years. Imaging examinations primarily utilized Philips platforms such as the iE33, GE HealthCare systems including Vivid E9, E95, or S5, and in one cohort Siemens equipment. Left ventricular ejection fraction was quantified either by the two-dimensional biplane Simpson method or volumetric three-dimensional echocardiography. When both modalities were available within a study, two-dimensional values were preferentially selected to reflect routine clinical practice and provide conservative estimates reflecting greater measurement variability.
Statistical modeling assumed a normal distribution of ejection fraction within healthy reference populations. Random-effects meta-analyses using restricted maximum likelihood were applied to pooled study means and log-transformed variances to derive overall pooled means and pooled standard deviations. Reference limits were calculated as the pooled mean plus or minus 1.96 times the pooled standard deviation, capturing the central 95 percent of the distribution. Z-scores and one-sided p-values compared against the lower 2.5th percentile were calculated for ejection fractions descending from 55 percent to 45 percent. Potential publication bias was evaluated through funnel plots and Egger regression testing, which revealed no statistically significant funnel plot asymmetry.
Analysis stratified by sex demonstrated that healthy women exhibit slightly higher systolic parameters than men. Among female participants, the pooled mean ejection fraction was 63.7 percent with a 95 percent confidence interval of 62.0 percent to 65.5 percent, a pooled standard deviation of 5.65 percent, and lower and upper reference limits of 52.7 percent and 74.8 percent, respectively. For male participants, the pooled mean ejection fraction was 61.9 percent with a 95 percent confidence interval of 59.8 percent to 64.0 percent, a pooled standard deviation of 5.19 percent, and corresponding lower and upper reference limits of 51.7 percent and 72.1 percent. Given these parametric distributions, fewer than 1 percent of healthy women and approximately 1 percent of healthy men would be expected to exhibit a measured ejection fraction below 50 percent.
When examining the influence of aging, meta-regression modeling across adult age brackets of 18 to 40 years, 41 to 65 years, and older than 65 years showed no statistically significant differences in mean ejection fraction for women (p = 0.71) or men (p = 0.94). Six included studies contributed age-stratified data. Although interstudy heterogeneity was observed, the researchers detected no consistent downward or upward trajectory across advancing adult age strata, indicating that an ejection fraction falling below 50 percent is unlikely to represent a benign age-related variation in younger or middle-aged adults, though adults aged 80 years and older were underrepresented.
Subgroup analyses by self-reported race and ethnicity identified modest differences, with participants of Asian origin demonstrating values two to three percentage points higher than White or Black cohorts. In Asian cohorts, mean values reached 64.9 percent in women and 63.6 percent in men, with lower normal limits of 53.8 percent and 53.2 percent, respectively. By comparison, White cohorts demonstrated mean values of 61.3 percent in women and 58.9 percent in men, while Black cohorts showed 61.5 percent in women and 60.4 percent in men, with lower normal limits of approximately 50 percent to 51 percent. Across all racial categories, the statistical probability that an ejection fraction below 50 percent represents normal cardiac function remained below 4 percent, while in Asian populations values below 54 percent in women and below 53 percent in men had less than a 2 percent probability of being normal.
Direct comparisons between two-dimensional and three-dimensional imaging cohorts revealed no systematic shift in pooled ejection fraction estimates, and machine manufacturer platforms did not account for differences across studies. However, the meta-analysis documented substantial statistical heterogeneity across cohorts, characterized by an overall between-study variance of tau-squared equal to 8.82 and an I-squared statistic of 99.7 percent. The authors noted that this high statistical heterogeneity was not explained by sex, age, or ultrasound modality, though true effect variation was partially mitigated within race-stratified groupings.
The study authors emphasize several methodological limitations that warrant cautious interpretation of the findings. Substantial heterogeneity likely stems from variable cohort inclusion criteria and differing body mass index cutoffs used to define healthy participants across original trials, alongside protocol differences in image acquisition. Although the between-study standard deviation of true effects was approximately 2.7 percentage points and individual cohort means clustered consistently between 60 percent and 69 percent without discordant outliers, residual uncertainty remains. Furthermore, Black and Hispanic adult populations were underrepresented in the available population-based literature, and the two-dimensional World Alliance Society of Echocardiography cohort was excluded because it reported only percentile-based boundaries without mean and standard deviation values. The authors conclude that while these findings reinforce current guideline thresholds near 52 to 53 percent and validate that an ejection fraction below 50 percent is abnormal, further research is required to evaluate high-risk demographics and elderly populations aged 80 and older.
Subject of Research: Normal left ventricular ejection fraction reference limits in healthy adults
Article Title: What is a normal left ventricular ejection fraction in healthy adults? A meta-analysis of population-based echocardiographic studies
Article References: Morsing, A. E., Gnesin, F., Lyass, A., & Andersson, C. (2026). What is a normal left ventricular ejection fraction in healthy adults? A meta-analysis of population-based echocardiographic studies. Journal of Cardiovascular Imaging, 34(1), Article 2. https://doi.org/10.1186/s44348-025-00063-4
Image Credits: AI Generated
DOI: 10.1186/s44348-025-00063-4
Keywords: left ventricular ejection fraction, echocardiography, transthoracic echocardiography, cardiology, heart failure, reference intervals, systolic function, meta-analysis, population-based studies, cardiac imaging, biplane Simpson method, three-dimensional echocardiography, sex differences, demographic reference ranges, American Society of Echocardiography, European Association of Cardiovascular Imaging, WASE study
News Source: Ophelia Keating. (November 19, 2026). Meta-Analysis Defines Normal Left Ventricular Ejection Fraction Limits in Healthy Adults. Scienmag.



