The National Cardiac Arrest Audit (NCAA) for the period 1/4/2015 – 31/3/2016 revealed a higher than expected number of cardiac arrests per 1000 hospital admissions at our Trust compared to national average (1). NCAA data for the period showed 44 cardiac arrests on hospital wards against an expected number of 30. The aim of our audit was to determine standards of care up to the cardiac arrest and potential for preventability.
Methods:
Multidisciplinary retrospective analysis of retrieved patients' notes. All notes reviewed separately and independently by ITU/Anaesthetics specialty doctor level or above and critical care outreach nurse. Case notes could be obtained for 38 of the 44 cardiac arrests.
Results:
Seventy-one percent (71%) of cases were > 70 yrs old and 42% of cases were >80 yrs old. DNACPR was not considered or discussed in 71% of cases. 21 cases out of 38 were from cardiac or respiratory ward or emergency assessment unit. The vital signs observation interval (as per hospital policy) before arrest was exceeded in 34% of cases. In 26% of patients the observations had been significantly overdue. Thirty-one cases (82%) showed room for improvement (clinical and/or organizational) according to NCEPOD classification (3).  4-5 cases (11-13%) were considered as probable or strong evidence of preventability according to Hogan score (2). Twenty two (58%) did not have return of spontaneous circulation (ROSC). Of the 16 patients with ROSC: 4 were admitted to ITU; 10 ward based ceiling of care; 2 did not require ITU. 3 patients (<8% of arrests) survived to discharge home (1 via ITU, 2 ward only) (national average: >18%). 60% of cardiac arrest calls happened during night shifts. More than 60% (23) cases were admitted to hospital >3 days before cardiac arrest. 27 of 38 cases had >2 organ failures/severe dysfunction upon time of admission. 15 cases had end stage disease (severe CCF with moribund status, severe COPD on home O2).
Discussion
Shortfalls in vital signs monitoring may have contributed to the number of cardiac arrests. In view of hospital length of stay before arrest, co-morbid status and proportion of night-time arrests it appears that mainly an appropriate treatment escalation plan had been lacking in many patients. This was also reflected in the high number of patients who after ROSC were deemed inappropriate for admission to ITU. The Trust has in the meantime moved to the national early warning score with a lower escalation threshold. The systematic use of treatment escalation plans should be considered to ensure appropriate patient selection.



References:
1-	Key statistics from the National Cardiac Arrest Audit 2015/16 – ICNARC.
2-	Hogan, H (2014) the Scale and Scope of Preventable Hospital Deaths. PhD thesis, London School of Hygiene & Tropical Medicine. DOI: 10.17037/PUBS.01776586.
3-	NCEPOD classification of care setting - Society for Acute Medicine, A SYSTEM TO MAXIMIZE LEARNING FROM MORTALITY REVIEWS ON AN ACUTE MEDICAL UNIT.  Meri Alvarez. 


08 / High in-hospital cardiac arrest rate in a district general trust