What is the clinical evidence from this study?

Study design: Observational. Population: mild initial COVID-19 disease (n=132). Intervention: Native T1 (CMR). Primary outcome: association between %-predicted VO₂peak and imaging parameters (β -0.25 per ms, 95% CI -0.40 to -0.10, p=< 0.001).

April 1, 2026Open Access

Native T1 is independently associated with aerobic exercise capacity in long-term follow-up after mild initial COVID-19 disease (Impression COVID&Heart Study)

Q: What are the key findings of this study?

Higher myocardial native T1 was independently associated with lower %-predicted VO₂peak (β = -0.25 per ms; 95% CI -0.40 to -0.10; p < 0.001) at 48 months follow-up after mild COVID-19.

Key Result

Higher myocardial native T1 was independently associated with lower %-predicted VO₂peak (β = -0.25 per ms; 95% CI -0.40 to -0.10; p < 0.001) at 48 months follow-up after mild COVID-19.

Study Design

Type

Observational (n=132)

Structured PICO

Population

132 individuals without prior structural heart disease evaluated at least 3 years after mild initial COVID-19 disease, mean age 49 ± 12 years, 52% male.

Outcome

Association between %-predicted VO₂peak (age-sex-BMI adjusted, SHIP reference) and imaging parameterssurrogate

Higher myocardial native T1 on CMR is independently associated with reduced long-term aerobic capacity in patients evaluated 4 years after mild COVID-19.

Main Result

Effect estimate: β -0.25 per ms (95% CI -0.40 to -0.10)

p-value: p=< 0.001

Abstract

BACKGROUND: Exercise intolerance is a common and incapacitating long-term consequence of COVID-19, even after mild acute illness. Cardiovascular magnetic resonance (CMR) studies have demonstrated persistent perimyocardial inflammatory abnormalities; however, their relationship with long-term aerobic capacity remains unclear. METHODS: In this prospective observational study, individuals without prior structural heart disease underwent standardised CMR, echocardiography, and cardiopulmonary exercise testing (CPET) at least 3 years after the initial COVID-19. The primary endpoint was the association between %-predicted VO₂peak (age-sex-body mass index (BMI) adjusted, Study of Health in Pomerania (SHIP) reference) and imaging parameters. Secondary analyses included lactate measurements and sex-stratified models. RESULTS: A total of 132 participants (mean age 49 ± 12 years; 68/132 (52%) male) were evaluated 48 months interquartile range (IQR) 42-53 post-infection. Non-ischaemic perimyocardial enhancement was present in 34/132 (26%), whereas two participants had an unrecognised ischaemic scar. Male sex, higher BMI, lower age, and higher native T1 were associated with lower %-predicted VO₂peak in univariate models. In multivariate analysis, male sex, lower age, and higher native T1 (β = -0.25 per ms, 95% CI -0.40 to -0.10; p < 0.001) remained independent predictors of lower %-predicted VO₂peak. In total, 63/132 (48%) participants demonstrated %-predicted VO₂peak below predicted values. Sex-stratified multivariate analyses showed that higher native T1 (p<0.001) independently associated with lower %-predicted VO₂peak in both men and women, with lower age additionally retained in men. In a lactate-measured subgroup (n=73), higher resting lactate, higher native T1, male sex, and lower left atrial area were associated with lower %-predicted VO₂peak. CONCLUSIONS: In long-term follow-up of individuals with mild initial COVID-19 and no prior structural heart disease, aerobic capacity relative to predicted values was reduced in 48% of participants, particularly in men, and was independently associated with higher myocardial native T1. Native T1 and resting lactate, but not conventional structural measures or peak exercise.