P73 Epidemiology and long term outcomes of patients with chronic obstructive pulmonary disease admitted to Scottish intensive care units

Abstract

Introduction: Patients admitted to intensive care units (ICUs) with an exacerbation of chronic obstructive pulmonary disease (COPD) are thought to have poor outcomes. Few studies have reported the impact of ICU admission on long-term outcomes. We aimed to describe the epidemiology of patients admitted to all Scottish ICUs with an exacerbation of COPD, and to ascertain the importance of admission on long-term mortality and hospital readmission rates. Method(s): We used a retrospective cohort design linking the Scottish Intensive Care Society Audit Group national database of ICU admissions to the national Scottish hospital discharges and death registry databases. The study cohort comprised all patients admitted to Scottish ICUs in 2005 aged =35 years with a primary ICU admission diagnosis of COPD, or respiratory infection/sepsis as the primary and COPD as the secondary diagnosis. Mortality was defined as death within 4 years of ICU admission. Logistic regression identified factors independently associated with mortality. Hospital readmission rates for the 3 years prior to and following discharge from the index hospitalisation were calculated. Result(s): 249 of 7727 (3.2%) first admission episodes to Scottish ICUs in 2005 were with an exacerbation of COPD. Mean age: 67 years; female: 52%; median admission Simplified Acute Physiology Score II: 35 (IQR 26-45). 83% of patients received mechanical ventilation; 24% required tracheostomy. Mean duration of ICU and hospital stays: 6.9 and 28.0 days, respectively. Mean duration of ventilation: 5.7 days (ICU survivors 5.9, non-survivors 5.2). Mortality at ICU and hospital discharge was 23.8% and 34.7%, respectively. It was 46.4% at 1 year and 66.1% at 4-year spost-ICU admission. Only age (p<0.001) and previous COPD admission (p=0.01) were independently associated with 4-year mortality. The mean hospital readmission rate during the 3-year follow-up period was twice that of the period before index hospitalisation (1.4 vs 0.7 readmissions/person-year for all admissions, 0.7 vs 0.3 for COPD admissions, Abstract P73 figure 1). (Figure presented) Conclusion(s): This Scottish national cohort had comparable shortterm mortality rates to those seen in a general ICU population. The poorer long-term mortality rate and increased readmission rates after hospital discharge may be useful to consider alongside other factors when deciding on ICU admission for patients with COPD.