In in vivo proton magnetic resonance (MR) spectroscopy, multiple quantum coherence (MQC) effects can give rise to unexpected modulations in the signal arising from J-coupled spin systems. When performing an in vivo MR spectroscopiс exam, the consequences of these modulations can be disastrous since improper pulse sequence timings can lead to the total disappearance of the signal of interest. The goal of this thesis was to evaluate both analytically and experimentally, the effects of MQCs on four clinically important J-coupled metabolites:​ lactate, an indicator of anaerobic respiration;​ citrate, a potential marker of adenocarcinoma in the prostate gland;​ Y-aminobutyric acid (GABA), the most prevalent neurotransmitter in the brain;​ and N-acetylaspartate (NAA), a measure of neuronal population in the adult brain.

When the resonance of interest exhibited minimal spectral overlap the analysis centred on the response of the metabolite to the commonly used STEAM and PRESS localization sequences. Equations were obtained that describe the response of the lactate doublet, the citrate multiplet and the aspartyl resonances of NAA to both of these sequences. In particular, the responses of both the citrate and NAA multiplets exhibited a strong magnetic field dependence arising from the effects of strong J-coupling.

In the case of GABA, a high degree of spectral overlap makes both the STEAM and PRESS sequences ineffective at typical in vivo field strengths. Тo observe GABA, selective MQC filtering sequences were designed that used static magnetic field gradient pulses to dephase unwanted coherences. The most effective editing of GABA was accomplished using reduced-duration, selective double quantum filters that gave excellent suppression of unwanted resonances while retaining a significant portion of the GABA triplet signal.

In all cases, the theoretical equations were derived by expanding the spin density operator for each spin system in terms of products of spin angular momentum operators, then following its evolution under a succession of time-independent Hamiltonians. 'The concise analytical expressions derived by this approach were in agreement with experiments performed on phantoms, rat brain extracts and in vivo human brain. The results may be used to predict the optimal pulse sequences and timings for the observation of each of the targetted metabolites.