Abstract 260: Patient Characteristics, In-hospital Mortality and the Use of Comfort-Focused Care among Patients presenting with Acute Ischemic Stroke

Abstract

Introduction: Little is known about the factors associated with the decision to shift care to comfort measure only (CMO) in patients admitted with ischemic stroke, and how they compare to patients who die despite resuscitation. We sought to determine factors associated with use of CMO and the short term outcomes in these patients. Methods: Using our institutional GWTG database, we analyzed 2,816 consecutive stroke admissions from 7/07 - 9/12 and classified patients into early CMO (<day 2), late CMO (≥ day 2) or non-CMO. Patient characteristics, in-hospital mortality and median (IQR) length of hospital stay (LOS) was compared across all groups. Results: Among 2,816 patients, 114 (4.0%) were made early CMO, 169 (6.0%) late CMO and 2,533 (90.0%) were never made CMO. In-hospital death occurred in 343/2816 (12.2%) overall and 283/343 (82.5%) of these cases were made CMO prior to death. 75.4% of early CMO, 71.6% of late CMO and 2.4% of non-CMO died during the hospital stay. There were substantial differences between baseline characteristics, in-hospital treatments, complications and median LOS in non-CMO patients compared with early and late CMO patients (Table). Comparing in-hospital deaths among the early plus late CMO cases (n= 207) vs. non-CMO (n=60) cases, there were no differences in baseline characteristics other than use of IA therapy (16.5% vs. 5.0%, P=0.02) and median (IQR) days from arrival to death [3 (2, 7) vs. 5 (3, 8), p=0.03]. Among early CMO patients, just over half died within the same day (28.7%) or 2nd day (26.4%) of becoming CMO, and ~25% survived past the day of discharge to hospice. Conclusion: While comfort-focused care was provided to only 10% of stroke admissions, it occurred in over 80% of all patients who died in-hospital. Early CMO patients received aggressive intervention with IA therapy less often and appear to have had fewer infectious complications, suggesting that poor initial neurologic prognosis and advanced directives may have contributed more than progressive deterioration to the decision to change care goals. Larger prospective studies are needed to better understand the drivers of decision-making in this population, and set realistic expectations about the likely duration of patient survival after CMO designation and the need for hospice services.