Background: Patients suffering in-hospital cardiac arrest (IHCA) often have abnormal clinical observations documented prior to the arrest. This study assesses whether these patients have a less favourable outcome following IHCA.
Methods: A multiple logistic regression analysis of retrospectively collected hospital chart data and prospectively collected Utstein style resuscitation data. Patients were defined as having abnormal clinical observations if they had one of the following documented 8 h before the arrest: systolic arterial blood pressure below 90 or over 200, pulse rate below 40 or over 140 beats per min or oxygen saturation below 90% with or without supplemental oxygen. Pre-arrest variables included were: age, sex and functional status, co-morbidities, reason for hospital admission, days in the hospital before the arrest, witnessed or un-witnessed arrest, arrest occurring outside regular working hours, monitored or non-monitored ward, whether basic life support was performed before the arrival of the resuscitation team, delay to arrival of resuscitation team and initial rhythm.
Results: Survival to hospital discharge of patients with clinically abnormal observations was 9% and among those without 18% (p=0.037). Independent pre-arrest predictors of survival were: un-witnessed arrest (odds ratio [OR] 0.1, confidence interval (CI) 0.01-0.8), initial rhythm other than ventricular fibrillation or ventricular tachycardia (OR 0.13, CI 0.05-0.3), delay to arrival of the resuscitation team exceeding 2 min (median) (OR 0.4, CI 0.15-0.9) and the presence of documented clinical abnormal observations prior to the arrest (OR 0.3, CI 0.09-0.95).
Conclusions: Patients with documented clinically abnormal observations before IHCA have a worse outcome than those without, despite prompt resuscitation. Efforts should be made to identify these patients in time, thereby possibly avoiding the arrest. This can also be used when assessing the prognosis in IHCA.