644

Abstract

Introduction: Patients with cancer may require intensive care (ICU) due to drug-related toxicity, organ dysfunction, or surgical complications. In the past, admission of cancer patients to the ICU was considered controversial due to the high mortality rates in this group of patients. More recently this effect appears to be diminishing possibly due to improvements in patient selection and supportive care. However, studies of patients with cancer include a large proportion of surgical patients in whom ICU outcomes are favourable. We therefore aimed to describe the ICU cancer population and to assess outcomes in cancer patients admitted to ICU with a medical diagnosis. Methods: We used routinely collected data from West of Scotland ICUs linked to Scottish Cancer registry to identify patients (aged 16 or over) admitted to ICU between 1st January 2000 and 31st December 2011. Patients who had a diagnosis of a malignant cancer (excluding non-melanoma skin cancer) within the previous 5-years were identified. We compared these patients to patients without cancer. Non-cancer patients had no diagnosis of cancer between 1984 and date discharged from ICU. Results: 49451 individual patients were admitted to an ICU during the study period. A malignant neoplasm had been diagnosed in 9704 (20%). The majority were diagnosed in the year prior to admission (7704, 85%). The number of admissions by major cancer type were as follows - colorectal 4223 (41% of admissions with cancer), stomach 692 (7%), lung 669 (6%), oesophageal 588 (6%) and bladder 425 (4%). 17% of cancer patients were admitted from a medical specialty compared to 50% of non-cancer patients. Of the patients admitted with a medical diagnosis, median age of patients with cancer was 65 (IQR 57-73) vs 56 (41-69) years for patients without cancer and median APACHE II score was 20 (9-27) vs 18 (10-25). The proportion of medical cancer and non-cancer patients in which organ support was provided was by invasive mechanical ventilation (71% vs 75%); vasoactive drug therapy (58% vs 46%); renal replacement therapy (20% vs 14%). The proportion of cancer patients with greater than one supported organ was 53% compared to 43% of patients without cancer. Mortality in ICU among medical cancer patients was 37.9% (95% CI 35.7 - 40.1%) compared to 25.8% (25.2 - 26.4%) in non-cancer patients. ICU mortality among cancer patients who did not receive organ support was 6.2% (1.7-15.0%) and 4.7% (3.3-6.4%) for non-cancer patients. Among patients who received three organ support mortality was 60.0% (45.1 - 73.6%) and 49.4% (44.9 -53.9%) among cancer and non-cancer patients respectively. Conclusions: Cancer patients admitted to ICU from a medical specialty tend to be older and have higher severity of illness scores. As a result there is a higher requirement for multi-organ support. Irrespective of the number of organs supported ICU mortality is higher among those patients with cancer.