T2 Relaxation Time Mapping of the Lumbar Spine at 3T MRI -Different Protocols Comparison

Aim: Different pulse sequences in magnetic resonance imaging have been developed for the quantification of T2 relaxation time of tissues. Quantitative T2 relaxation time measurement is proposed to be valuable in detecting the early sign of disc degeneration. The purpose of this study was to evaluate the consistency of T2 relaxation time quantification in the spine tissues, including disc, bone and fat using different sequences and to assess the MultiTransmit technology effect on T2 quantification at 3.0T MR scanner. Materials and Methods: T2 relaxation times were first measured on eleven commercial phantoms with different T2 relaxation time in the range from around 50ms to 150 ms. Pulse sequences used for the measurement of the T2 relaxation time included with the traditional spin echo (SE), multi spin echo, turbo spin echo with various echo train length (ETL), gradient and spin echo (GRASE) sequences with single-slice on 3.0 T MR unit. Same series of pulse sequences were applied to lumbar spine of 15 healthy volunteers. T2 relaxation times of disc, vertebral body and fat were then measured. In addition, the B1 effect on T2 quantification was assessed by switching on or off the multi-transmitter (MT) of the MR scanner. Coefficients of variation (CV) of T2 from each sequence were calculated for reproducibility assessment. The differences between these T2 measurements from each sequence were statistically analysed using student's t test. Results: A good reproducibility in all five phantoms was demonstrated by the CVs of T2 measurement which was lower than 4% for all sequences. The CVs is lower for MT switching on than those for MT switching off for all phantoms. The T2 relaxation times obtained from protocols with MT switching off were higher than those obtained from the same protocols with MT switching on in all phantoms and fat tissue except for the SE protocols in spine discs and marrows. There is significant difference (p<0.02) for all sequences in both phantoms and healthy volunteers obtained with MT switching on from with MT switching off. With MT switching off, the average T2 relaxation time was 127.26±18.94ms (ranging from 95ms to 170ms) in discs, 88.10±28.34ms (ranging from 41ms to 157ms) in vertebral bone and 132.97±46.69ms (ranging from 39ms to 194ms) in subcutaneous fat respectively. With MT switching on, the average T2 relaxation time was 121.22±15.27ms (ranging from 93ms to 159ms) in discs, 83.04±27.96ms (ranging from 41ms to 145ms) in vertebral bone and 126.09±49.19ms (ranging from 41ms to 196ms) in fat respectively. The times for fat and marrow varied more for both MT switching on and off. The SE sequence yielded the lowest T2 values for all in vivo tissues and phantoms, which was significantly lower than those obtained from other sequences (P<0.006). In TSE protocols, T2 relaxation time increases as ETL increases and there is significant difference (P<0.05) whatever MT switching on or off. T2 values from GRASE sequence were lower than those from TSE protocols but the difference were not significant (P>0.05). Conclusion: T2 mapping was viewed as one of the valuable methods for the diagnosis of intervertebral disc degeneration. There are different pulse sequences for the T2 relaxation time measurement. It is more appropriate to compare the trends of relaxation times rather than the actual relaxation time especially for comparing studies because many factors affect T2 values.