Optimizing T1rho and T2 Values for Intervertebral Discs Obtained from the Combined T1rho and T2 Sequence

Abstract

To investigate the effect of the number of spin-lock (SL) and T2 preparation pulse using T1rho and T2 values obtained from the combined T1rho and T2 sequence. We included 30 patients who underwent magnetic resonance imaging of the lumbar spine because of low back pain and leg numbness, tingling, and pain. We used 3D turbo-field echo and the adiabatic pulse as SL pulse for T1rho mapping and the block pulse as T2 preparation pulse for T2 mapping of the combined T1rho and T2 sequence. The preparation time of T1rho and T2 was set at 0, 20, 40, 60, and 80 ms. We defined the T1rho and T2 values calculated from all SL and T2 preparation pulses as Dfull and decreased several number of SL and T2 preparation pulses from Dfull as other groups (D1, D2, and D3). We used the Bland–Altman analyses to estimate the systematic and proportional bias between Dfull and other groups. The 95% CI of the mean difference included zero in all groups. Therefore, systematic bias was not detected. The regression coefficients with D3 of the T1rho and T2 value were − 0.34 and − 0.23, respectively (p < 0.01). We detected the proportional bias in the T1rho and T2 values in only D3 (0 and 80 ms). An investigation of the T1rho and T2 values of IVDs using the combined T1rho and T2 sequence suggested that the accuracy of these values decreased with suitably adjusted three preparation pulses, facilitating the assessment of both T1rho and T2 values at approximately 10 min.