Optimization of pulse sequences for lactate detection using MR spectroscopy

Lactic acid is one of metabolisms from the glycolysis. Brain anoxia and ischemia will cause accumulation of lactate. Noninvasive detection of lactate using MR spectroscopy has played an important role in diagnosis and treatments of diseases. But lactate is usually difficult to be detected in most of cases. The purpose of present study was to obtain optimal detection and quantitative analysis of lactate by changing a series of parameters used in the sequences. The experiment was carried out in a glass model containing lactate and 12 patients with brain diseases. The MR scanner is a 1.5T system (Signa MR/i, GE). The spectra were obtained by using both PRESS and STEAM sequences. The spectra were processed using a GE software. The curves of lactate peaks versus TE obtained by the model exhibited waveform characteristic. The up doublets were recorded with their maximum peak heights using TE values of 26 ms and 270 ms, whereas the down doublets were recorded at their peaks using a TE value between 135 to 160 ms. All the peaks recorded by STEAM sequence were weaker than those by PRESS sequence. The results indicate that the optimized sequences are useful for better detection of lactate in clinical examinations.