Patients expected to develop life-threatening complications in acute meningococcal infections require early recognition and appropriate monitoring. Different prognostic scoring systems have been developed. Three of them, chosen according to their bedside availability, were compared with our clinical observations. Twenty consecutive cases of proven meningococcal infection were admitted to the paediatric intensive care unit (PICU) of the Free University of Brussels (AZ-VUB). Biological and clinical features required for prognostic scoring were evaluated as soon as possible after admission. Glasgow meningococcal sepsis prognostic score (GMSPS), Neisseria sepsis index (NESI) and Algren criteria were retrospectively calculated and evaluated for their prognostic significance. Neisseria meningitidis was cultured from blood and cerebrospinal fluid in 11 patients and from blood in only nine patients. The age of the patients was between 1 and 15 years (mean 4.1 years). All patients received the same therapy on admission. Four patients died with a multiorgan failure within 18 hours. The three scoring systems in these four patients predicted death. Overall, the GMSPS score, the NESI score and the Algren criteria predicted death in respectively 10, nine and five patients. Death was falsely predicted in six patients by the GMSPS score, in five patients by the NESI score and in one patient by the Algren criteria. The Algren criteria predicted the severity of the clinical process more accurately than did the GMSPS and NESI scores. However, such predictability should be cautiously used even when 100% mortality is predicted. It might be used in decision-making in regard to the following issues: patient transfer to tertiary centres and mode of transportation, monitoring of patients in intensive care units, early insertion of invasive cardiovascular monitoring catheters and consideration of new or even experimental therapy. However, one should be extremely cautious of taking any therapeutically or ethical decision on the basis of one or more of the described scoring system, since we showed the lack of precision concerning the outcome of paediatric patients with meningococcaemia.