Giant cell arteritis (GCA) is the prime medical emergency in ophthalmology because of its dreaded complication of visual loss in one or both eyes, which is preventable if these patients are diagnosed early and treated immediately and aggressively with systemic corticosteroids. However, there is much controversy on diagnostic criteria and various aspects of steroid therapy to prevent visual loss. We discuss in detail the reasons for the controversy, clinical criteria to establish a definite early diagnosis of GCA, and its management. To provide new information on corticosteroid therapy in GCA, we also present our 27-year planned study on steroid therapy in GCA in 145 temporal artery biopsy-confirmed GCA patients (96 with and 49 without visual loss) seen and followed for 6 weeks or more in our clinic. The median follow-up time was 2.43 years, with interquartile range of 1–6 years (range 6 weeks to 20.2 years). Intravenous megadose steroid therapy was initially given to 33% followed by oral steroids, while the rest had only the oral therapy. The median starting oral prednisone dose was 80 mg/day, with 40% on ≧100 mg/day. We found that the most reliable and sensitive parameters to regulate and taper down steroid therapy were the levels of erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) and not systemic symptoms. All patients were maintained at the high-dose prednisone till both the ESR and CRP had stabilized at low levels (that usually took 2–3 weeks), after which very gradual tapering of prednisone was started, guided by the ESR and CRP levels only. The median time to reach the lowest maintenance dose of prednisone at which the ESR and CRP stayed low and stable was 48.7 months (95% CI: 34.6, 71.4 months), and the median lowest prednisone dose achieved was 7 mg/day (interquartile range of 1–16 mg/day). A comparison of patients with and without visual loss showed no significant difference in the time to attain the lowest dose (p = 0.359). Our study showed that no generalization is possible for tapering down of prednisone and there is no set formula because of the infinite variation between individuals. Only 10 (7 without visual loss, 3 with visual loss) of 145 patients were able to stop the therapy and maintain stable ESR and CRP levels. We found that only 4% of GCA patients with visual loss showed any visual improvement with high-dose steroid therapy, and 4% developed further visual loss during the first 5 days of high-dose steroid therapy but none after that. Our studies found no evidence that intravenous megadose steroid therapy was more effective than oral therapy in improving vision or preventing visual deterioration due to GCA.