Three-dimensional SPACE MR with CAIPIRINHA fourfold acceleration for assessing long head of biceps tendon

Background Data regarding controlled aliasing in parallel imaging results in higher acceleration (CAIPIRINHA) T2-weighted sampling perfection with application optimized contrast evolution (SPACE) with fourfold acceleration factor for assessing long head of biceps tendon (LHBT) disorder is lacking. Purpose To investigate the feasibility of 3D CAIPIRINHA SPACE with fourfold acceleration in assessing LHBT disorder. Material and Methods A total of 42 consecutive patients underwent shoulder magnetic resonance (MR) examinations including CAIPIRINHA SPACE with fourfold acceleration, and non-CAIPIRINHA SPACE with twofold acceleration, and 2D fast spin echo (FSE). A subjective score of depiction of LHBT was given to 3D sequence according to a 4-point scale (0-3, "poor" to "excellent"). The Wilcoxon signed rank test was used to compare depiction scores between 3D sequences. Three statuses of LHBT were defined in the study: normal, tendonitis, and tear. McNemar's test was used compare diagnostic accuracy. Results LHBT was better depicted with CAIPIRINHA SPACE versus non-CAIPIRINHA SPACE (2.1 +/- 0.4 vs. 1.5 +/- 0.4; P < 0.001). Inter-modality agreement between CAIPIRINHA SPACE and 2D FSE was almost perfect (kappa = 0.884 +/- 0.064). The sensitivity and specificity in detecting LHBT disorder were 95% (20/21) and 95% (20/21), respectively, for CAIPIRINHA SPACE, and 71% (15/21) and 76% (16/21), respectively, for non-CAIPIRINHA SPACE (P = 0.039). Conclusion Fourfold acceleration CAIPIRINHA is feasible in reducing the acquisition time of SPACE MR in the shoulder. 3D CAIPIRINHA SPACE with fourfold acceleration is highly accurate in detecting LHBT disorder.