Acute renal failure (ARF) is at a crossroads between nephrology and intensive care medicine. However, there seems to be wide differences between the ARF observed in the intensive care unit (ICU) compared to that observed in other areas of the hospital, particularly when examining the mortality rate. Among the ICU patients the 70% mortality rate is higher to the 50% found in an overall series of studies. Recently, Druml proposed that there is a changing trend in the clinical spectrum of ARF as a convincing reason to justify these differences. According to him, we are moving from an ARF seen as a mono-organ failure to another one observed in a multiorgan dysfunction syndrome (MODS) context. Although extremely coherent, this hypothesis has not been fully confirmed in a prospective study. In fact, most authors seem to look at the problem from opposite sides of the river, either from the critical medicine or the nephrological bank. To the best of our knowledge, only one retrospective study has dealt with this topic by comparing outcome of ARF in ICU and non-ICU patients. In this article we aim to overcome this problem by reviewing the data of the prospective epidemiological ARF study carried out in Madrid using two different approaches: (1) comparing the ARF cases observed in the ICU setting with those ARF studied outside the ICU, and (2) comparing the outcome of isolated ARF with the outcome of ARF as part of a MODS in patients treated in both settings.