Age- and sex-specific reference ranges for cardiac function and structure in Germany: cardiovascular magnetic resonance imaging (CMR) in the German National Cohort (NAKO)
Authors
- Christopher L. Schlett
- Christopher Schuppert
- Peter M. Full
- Robin T. Schirrmeister
- Manuel Hein
- Marco Reisert
- Maximilian F. Russe
- Martyna Flis
- Jan Gröschel
- Clemens Ammann
- Valeria Geiger
- Karin H. Greiser
- Tafirenyika Gwenzi
- Anna Kottgen
- Thomas Kröncke
- Thomas Küstner
- Wolfgang Lieb
- Lea Jigme Michel
- Konstantin Nikolaou
- Annette Peters
- Tobias Pischon
- Henning Teismann
- Henry Völzke
- Klaus H. Maier-Hein
- Fabian Bamberg
- Susanne Rospleszcz
- Jeanette Schulz-Menger
Journal
- medRxiv
Citation
- medRxiv
Abstract
INTRODUCTION: Cardiovascular magnetic resonance (CMR) is the reference standard for quantifying cardiac structure and function, yet widely applicable population-based reference values remain limited. We derived age- and sexspecific reference ranges for ventricular volumes, mass, and function using data from the population-based German National Cohort (NAKO). METHODS: Short-axis balanced steady-state free precession cine images from 3T CMR of 29,908 participants were analyzed using a validated deep learning segmentation pipeline with systematic quality control. From these, we defined a main reference cohort free of cardiovascular disease (CVD), and a healthy subcohort additionally free of cardiovascular risk factors. Standard left (LV) and right ventricular (RV) measures were quantified and indexed. Reference intervals (5th–95th percentiles) were modeled using additive models and quantile regression to capture non-linear age trends, stratified by sex, with formal testing for age-sex interactions. RESULTS: The CVD-free reference cohort included 24,371 participants (mean age 43.8±12.2 years; age range 20-72 years, 44.7% women). LV and RV end-diastolic and end-systolic volumes declined with age, whereas LV ejection fraction remained stable and RV ejection fraction increased modestly. Sex differences were consistent across metrics and all major parameters demonstrated significant age-sex interactions; differences were most pronounced at younger ages and attenuated in later life. The healthy subcohort (n=5,550) showed similar structural and functional profiles, without clinically relevant deviations from the main reference cohort. CONCLUSIONS: This study provides age- and sex-specific CMR reference ranges derived from a large, uniformly imaged national cohort. These data offer a population-based normative framework for clinical CMR interpretation and future research on sex-specific cardiac remodeling and healthy aging.