Globally there are ≈4.9 million bilaterally corneal blind and 23 million unilaterally corneal blind. Majority of this blindness exists in the developing countries, where resources for corneal banking and transplant surgery are less than adequate. Survival of corneal grafts gradually declines over the long term. Corneal transplantation has poor prognosis in vascularised corneal beds, ocular surface disease and viral keratitis. Keratoprosthesis (KPro) remains as a final option for end-stage ocular surface disease, multiple corneal transplant failures and high-risk corneal grafts. Boston type-1 KPro and osteo-odonto-keratoprosthesis are the two devices proven useful in recent years. Choice of a keratoprosthetic device is patient specific based on the underlying diagnosis, ocular morbidity and patient suitability. KPro surgery demands a high level of clinical and surgical expertise, lifelong commitment and extensive resources. Improvements in techniques and biomaterials may in the future provide retainable KPros that do not need regular follow-up of patients, have low complications but high retention rates and may be produced at a low cost on a mass scale to be available as 'off the shelf' devices. Because KPros have the potential to effectively address the burden of surgically treatable corneal blindness and may also eliminate the problems of corneal transplantation, more research is required to develop KPros as substitutes for corneal transplantation even in low-risk cases. In those countries where corneal blindness is a major liability, we need a two pronged approach: one to develop eye donation, eye banking and corneal transplantation and the second to establish centres for keratoprostheses, which are affordable and technically not challenging, in a population where default on follow-up visits are high. Until the latter is achieved, KPros should be viewed as a temporary means for visual restoration and be offered in national and supraregional specialised centres only.