OBJECTIVES: Cardiopulmonary resuscitation (CPR), a life-saving technology developed mid-20th century, was intended to prevent sudden, unexpected death. Since then, CPR has increasingly been performed on terminal patients with irreversible illness treated in acute care settings. The objectives were to: synthesize the literature estimates of survival following CPR delivered in acute care hospitals; identify predictors of survival; and contextualize the findings in terms of ethics of end of life care. METHODS: We conducted a comprehensive literature review using PUBMED to identify studies in which survival after CPR was reported for adult patients treated in hospitals. RESULTS: The literature yielded 130 studies reporting survival after CPR. The most common outcomes were return of spontaneous circulation (52% of studies, arithmetic mean survival 43% (range 12% to 100%)) and survival to hospital discharge (64% of studies, arithmetic mean survival 15% (range 0% to 40%)). Diagnoses with lower rates of survival included metastatic cancer, sepsis, renal failure, pneumonia, stroke, diabetes and AIDS. CONCLUSIONS: While heterogeneity in study populations and outcomes invalidated summary estimates, all studies showed that only a minority of patients undergoing CPR survived even to hospital discharge. Current use of CPR in many hospitalized patients likely does not achieve consensus goals, to: preserve life, restore health, and limit disability. Reviving terminally ill patients or those with major co morbidities should involve consideration of the efficacy, preferences of patients and caregivers, and burdens imposed. However, unlike Prometheus who was punished for stealing fire from Zeus and handing it to mankind, the burden of CPR is carried by patients undergoing CPR and by those left untreated through spent opportunity costs. This serves as a reminder that the technological imperative – because existing technology means that something can be done, then this action ought to be done – must be thoughtfully considered by those undertaking health technology assessments.