Magnetic Resonance Spectroscopy as a Non-invasive Method to Quantify Muscle Carnosine in Humans: a Comprehensive Validity Assessment

Abstract

Carnosine is a dipeptide abundantly found in human skeletal muscle, cardiac muscle and neuronal cells having numerous properties that confers performance enhancing effects, as well as a wide-range of potential therapeutic applications. A reliable and valid method for tissue carnosine quantification is crucial for advancing the knowledge on biological processes involved with carnosine metabolism. In this regard, proton magnetic resonance spectroscopy (1H-MRS) has been used as a non-invasive alternative to quantify carnosine in human skeletal muscle. However, carnosine quantification by 1H-MRS has some potential limitations that warrant a thorough experimental examination of its validity. The present investigation examined the reliability, accuracy and sensitivity for the determination of muscle carnosine in humans using in vitro and in vivo experiments and comparing it to reference method for carnosine quantification (high-performance liquid chromatography – HPLC). We used in vitro 1H-MRS to verify signal linearity and possible noise sources. Carnosine was determined in the m. gastrocnemius by 1H-MRS and HPLC to compare signal quality and convergent validity. 1H-MRS showed adequate discriminant validity, but limited reliability and poor agreement with a reference method. Low signal amplitude, low signal-to-noise ratio, and voxel repositioning are major sources of error.