OBJECTIVES: To assess how perfusion interval relates to myocardial injury and whether this is modified by aortic cross-clamp (ACC) duration in minimally invasive mitral valve surgery. METHODS: We retrospectively analysed 556 patients receiving blood-enriched modified del Nido cardioplegia. The primary exposure and outcome were longest perfusion interval and base-10 logarithm-transformed peak cardiac troponin I (cTnI) within 48 hours. Analyses included multivariable regression, interaction and subgroup analysis, restricted cubic splines, sensitivity analysis, and logistic regression. RESULTS: Longer perfusion intervals were initially associated with higher cTnI (regression coefficient 0.0097, 95% confidence interval [CI], 0.0035-0.0159; P = .002). This association disappeared after adjustment for factors including ACC time and number of perfusions (regression coefficient -0.0006, 95% CI, -0.0109 to 0.0097; P = .913). Aortic cross-clamp time remained independently associated with cTnI (regression coefficient 0.0099, 95% CI, 0.0013-0.0185; P = .024). In patients with prolonged ACC time, exploratory analysis suggested a non-linear increase beyond 80 minutes. In patients with ACC time of 90 minutes or less, perfusion interval was not clearly associated with peak cTnI. No significant associations were found with the composite of adverse events. CONCLUSIONS: The association between perfusion interval and myocardial injury depended on ACC duration. Single-dose blood-enriched modified del Nido cardioplegia appeared adequate when ACC time was 90 minutes or less. Longer intervals in prolonged procedures may matter, but the 80-minute threshold remains exploratory. These findings support a context-dependent myocardial protection strategy.