AIM: To evaluate the clinical presentations of traumatic optic neuropathy and to assess the visual outcome of three groups of patients managed differently (conservative, intravenous corticosteroids only and combination of intravenous and oral corticosteroids) at an academic tertiary care referral centre. METHODS: A retrospective study was conducted involving 24 consecutive patients (27 eyes) with traumatic optic neuropathy attending Hospital Universiti Sains Malaysia from January 2007 till December 2009. RESULTS: Twenty-four patients (27 eyes) were included. All cases involved were males. Mean age was 33 years old. Motor vehicle accident was the major cause (83.3%). Both eyes were equally involved. Most of the eyes had poor vision on presentation (HM-NPL, 81.5%) with associated periorbital haematoma (22 eyes) and subconjunctival haemorrhage (20 eyes). Majority of patients (19 patients, 79.2%) presented with more than one bony fracture of skull or orbit and 5 patients (20.8%) had no fractures. None of the patients had evidence of optic nerve compression on CT scans or MRI done. Eleven patients (45.8%) had been treated with intravenous and oral corticosteroids. The other 7 patients (29.2%) were treated conservatively and the third group (6 patients, 25.0%) was on intravenous corticosteroids only. Eleven of 12 eyes (91.7%) treated with intravenous and oral corticosteroids had shown 1 line improvement of visual acuity. Those eyes treated conservatively (77.8%) had shown 1 line improvement of visual acuity. As for patients treated with intravenous corticosteroids only, four patients remained NPL, one patient had mild visual improvement and the other one's vision remained the same. The visual improvement in patients treated with conservative management was not significant (P=0.386). Patients treated with intravenous corticosteroids alone have shown no visual improvement statistically(P<0.05). Patients treated with intravenous followed by oral corticosteroids had significant visual improvement (P<0.05). There was no statistically significant difference in visual outcome between patients treated with corticosteroids and patients treated conservatively (P=0.368). No patient underwent surgical decompression of the optic nerve. In this series, the follow up ranges from 6 months to 3 years. CONCLUSION: Most of the traumatic optic neuropathy patients presented with periorbital haematoma, subconjunctival haemorrhage and orbital wall fractures. Patients treated with intravenous followed by oral corticosteroids have better visual outcome compared to conservative management.