Very old patients urgently referred to the intensive care unit: long-term outcomes for admitted and declined patients.

BACKGROUND Outcomes for very old patients (≥ 80 years) referred but not admitted to an intensive care unit have not been described in Australia and New Zealand. OBJECTIVE To ascertain long-term (12-month) outcomes for very old patients urgently referred for ICU support at a tertiary referral hospital. DESIGN, SETTING AND PATIENTS A retrospective, medical record review of 1240 very old patients (≥ 80 years) who were urgently referred to an Australian, 31-bed ICU over a 40-month period from March 2011 to August 2014. Referrals were divided into those who were "too well" for the ICU, admitted to the ICU, and "too sick" for the ICU. Data and main outcome measures: Data were extracted from hospital records, the ICU patient database and the Australian Institute of Health and Welfare National Death Index, and our main outcome measures were health status and destination at hospital discharge, and 12-month mortality rates. RESULTS Urgent admissions of very old patients accounted for 6.9% of total ICU admissions (443/6415). The hospital mortality rate was 16.0% (93/583) for patients who were too well, 32.1% (142/443) for those admitted to the ICU, and 69.2% (148/214) for those too sick (P ≥ 0.001). Mortality rates 12 months after referral were 40.8% (238/583), 46.0% (204/443) and 88.3% (189/214), respectively (P ≥ 0.001). CONCLUSION Very old patients considered too well for the ICU have a significantly lower hospital mortality rate than those admitted to the ICU after urgent referral. However, 12 months after referral, patients considered too well for ICU admission have a mortality rate approaching that of very old patients admitted to the ICU. Over half of very old patients urgently referred to the ICU die within 12 months.