Optimization of Respiratory Training Methods for Cardiac Magnetic Resonance Imaging

BACKGROUND: In cardiac MRI, poor breath-holding may lead to suboptimal image quality. However, the effect of different breath-hold methods on image quality remains unknown. PURPOSE: To explore the effect of respiratory training with different breath-holding methods and identify the optimal technique for reducing respiratory motion artifacts in cardiac MRI. STUDY TYPE: Prospective. POPULATION: A total of 93 patients undergoing cardiac MRI (64 men and 29 women; mean age of 48.9 ± 13.7 years). FIELD STRENGTH/SEQUENCES: 3T, single-shot fast spin-echo sequence, fast imaging employing steady-state acquisition sequence, first-pass perfusion sequence (fast gradient echo), and late gadolinium enhancement sequence (phase-sensitive myocardial delayed enhancement). ASSESSMENT: All patients undergoing cardiac MRI were divided into three groups: mouth-breathing, nasal-breathing, and nose-clip. Respiratory-gated monitoring diagrams, examination duration, and subjective and objective image quality assessments were compared. STATISTICAL TESTS: One-way ANOVA, Chi-square test, Kruskal-Wallis H test, Kappa test. Statistical significance was set at p < 0.05. RESULTS: Evaluation of the cardiac MRI respiratory-gated monitoring diagrams showed that the probability of good breath-holding in the nose-clip group was 96.7%, significantly higher than in the mouth-breathing (64.5%) and nasal-breathing groups (75.0%). There was no significant difference in examination duration among groups. Myocardial noise in the mouth-breathing group was significantly higher than those in the other two groups (24.2 [18.3, 32.5] vs. 15.8 [12.5, 23.9]/19.2 [13.7, 24.4]); blood pool noise was significantly higher than in the nasal-breathing group (34.6 [22.6, 42.9] vs. 24.9 [18.4, 33.8]). However, CNR did not significantly differ among the three groups (p = 0.07). Subjective image quality scores were higher in the nose-clip group than those in the other two groups (26 [86.7%] versus 16 [51.6%]/18 [56.3%]). DATA CONCLUSION: Optimized respiratory training can enhance breath-holding performance and image quality in patients undergoing cardiac MRI. The use of a nose clip for assisted breath-holding was found to be the most effective method. EVIDENCE LEVEL: 2. TECHNICAL EFFICACY: Stage 2.