Background: Recent basic and clinical evidence suggests amiloride may be neuroprotective in multiple sclerosis (MS) through the blockade of the acid sensing ion channel (ASIC). Objective: To examine the neuroprotective efficacy of amiloride in acute optic neuritis (ON). Methods: A total of 48 patients were recruited to a phase 2, double blind, single site, randomised controlled trial. Scanning laser polarimetry (GDx) at 6 months was the primary outcome measure and optical coherence tomography (OCT) and visual and electrophysiological measures were secondary outcome measures. Participants aged 18–55 years, ≤28 days of onset of first episode unilateral ON, were randomised to amiloride (10 mg daily for 5 months) or placebo ( clinicaltrials.gov , NCT 01802489). Results: Intention-to-treat (ITT) cohort consisted of 43 patients; 23 placebo and 20 amiloride. No significant drug-related adverse events occurred. No significant differences were found in GDx ( p = 0.840). Visual evoked potentials (VEP) were significantly prolonged in the amiloride group compared to placebo ( p = 0.004). All other secondary outcome measures showed no significant difference. Baseline analysis of OCT data demonstrated a significant pre-randomisation thinning of ganglion cell layer. Conclusion: Amiloride has not demonstrated any neuroprotective benefit within this trial paradigm, but future neuroprotective trials in ON should target the window of opportunity to maximise potential neuroprotective benefit.