Introduction In vivo 31P Magnetic Resonance Spectroscopy (MRS) measures phosphorus-containing metabolites that play an essential role in many disease processes. An advantage over 1H MRS is that total choline can be separated into phosphocholine and glycerophosphocholine which have opposite associations with tumour grade. We demonstrate 31P MRS can provide robust metabolic information on an acceptable timescale to yield information of clinical importance. Methods All MRI examinations were carried out on a 3T whole body scanner with all 31P MRS scans conducted using a dual-tuned 1H/31P head coil. Once optimised on phantoms, the protocol was tested in six healthy volunteers (four male and two female, mean age: 25 ± 2.7). 31P MRS was then implemented on three children with optic pathway gliomas. Results 31P MRS on volunteers showed that a number of metabolite ratios varied significantly (p < 0.05 ANOVA) across different structures of the brain, whereas PC/GPC did not. Standard imaging showed the optic pathway gliomas were enhancing on T1-weighted imaging after contrast injection and have high tCho on 1H MRS, both of which are associated with high grade lesions. 31P MRS showed the phosphocholine/glycerophosphocholine ratio to be low (<0.6) which suggests low grade tumours in keeping with their clinical behaviour and the histology of most biopsied optic pathway gliomas. Conclusion 31P MRS can be implemented in the brain as part of a clinical protocol to provide robust measurement of important metabolites, in particular providing a greater understanding of cases where tCho is raised on 1H MRS.