Does MRS Lactate Peak Correlate with Lactate in the CSF and Blood?

Abstract

Purpose Cerebrospinal fluid (CSF) or brain parenchyma lactate detection is important for the diagnosis of some diseases with aerobic cellular metabolism compromise. Our purpose is to correlate intraventricular magnetic resonance spectroscopy (MRS) lactate detection and quantification to CSF and blood lactate concentration. Methods Twenty-one patients (13 females; mean age 5 years) suspected of having mitochondrial disorders underwent proton MRS with point-resolved spectroscopy (TE = 144 ms). The volume of interest was positioned in the lateral ventricles, and LCModel was used for the MRS lactate peak detection and quantification. CSF and venous blood samples were obtained for lactate quantification immediately after MRS. Comparisons between MRS, CSF, and blood lactate detection and quantification were performed. p < 0.05 was considered significant. Results In our series, CSF lactate levels were high in 11 patients (52%) and blood serum lactate levels were high in 3 patients (14%). MRS was able to detect a lactate peak in all patients. A positive correlation between MRS lactate quantification and CSF lactate was observed (Pearson correlation coefficient = 0.750; p < 0.0001). Blood lactate did not correlate with lactate levels in the lateral ventricle measured by MRS or direct determinations of CSF lactate concentration. When noting the lactate quantification obtained by MRS, only 8 patients (as opposed to 14) presented higher values than the established normality threshold of 1.7 mmol/L. Considering CSF puncture concentration as the gold standard, we obtained 64% of sensitivity and 90% of specificity for MRS quantification. Conclusion If MRS shows increased lactate levels in the ventricles, CSF puncture is not needed for lactate increase confirmation.