Intensive care is increasingly being used in the management of cancer patients. It is important that a disproportionate share of special care resources is not expended on futile care of terminally ill patients. A requirement for mechanical ventilation has been stated to affect survival in cancer patients. The objectives of this study were to determine our hospital utilisation of ICU facilities and the prospects of a successful outcome in cancer patients with a need for ventilatory support. The Norwegian Radium Hospital is a 400-bed cancer hospital with a 12-bed combined postoperative and intensive care unit (PO/ICU). For each patient admitted to the PO/ICU, patient data including diagnosis, therapeutic interventions, use of resources and outcome are entered in a computerised database. We reviewed all 10,051 patients admitted during a 5-year period, focusing on the patients receiving ventilatory support. There were 347 patients who were treated with mechanical ventilation, 228 patients only for a short period postoperatively after extensive surgery. A further 119 patients (mean age 68 years, mean SAPS 33.5) were treated with mechanical ventilation for more than 24 h or died during treatment in the ICU; 65 patients (55%) were admitted after elective surgery, 24 (20%) after surgical emergencies and 30 (25%) after medical emergencies. Metastatic disease was present in 59% of them. These 119 patients comprised 1.18% of all patients admitted to the PO/ICU, but utilised 28% of all resources. They included 34 patients (29%) who died during the ICU stay, while 69 patients (58%) were still alive after 6 months. The ICU mortality in different groups was: surgical patients 24%, gynaecological patients 9%, oncological patients 63%. The mortality in the age group >70 years was 15%. The role of ICU facilities, including mechanical ventilation, is important for optimal supportive care in cancer patients. Our results indicate that this treatment modality should not generally be restricted in critically ill cancer patients. The quality of life of the patients who survived should be of interest to those involved in further medical and ethical decisions concerning the level of care in the ICU.