Patient monitoring in Intensive Care Units (ICUs) requires the acquisition of a large amount of information and/or data. This information will be later analyzed and interpreted by clinical specialists. The task becomes particularly crucial when it involves intensive care units that incorporate a large number of beds. Most engineering efforts in the field have followed two different approaches: 1) Construction of systems and mechanisms for massive data acquisition, storage and processing of data, and appropriate display of the results; and 2) Building expert systems. This paper presents the basis for an alternative approach which combines deterministic and heuristic techniques. The approach also takes into account the possibility of reasoningunder uncertainty; one of the intrinsic characteristics of the clinical inferential processes.