Results from inhospital cardiopulmonary resuscitation records in a medical cardiologic ICU

Introduction

New techniques have been used in cardiopulmonary resuscitation (CPR) since the introduction of closed cardiac massage in 1960. Despite this progress, there has been no significant improvement in survival rates after inhospital cardiac arrest over the past 40 years. In a general hospital, survival rates at discharge, not considering specifically ICU patients, is around 15–20%. Few data are available considering survival in cardiologic care units.

Methods

Between April 2004 and December 2004 we recorded and analysed all attempted cardiopulmonary resuscitation in a medical ICU of a teaching cardiologic hospital. The patients were 64 ± 20 years old, 44 (62.8%) male and 26 (37.2%) female. Diagnosis at admission in ICU were: cardiogenic and/or septic shock 37.2%, heart failure (NYHA IV) 20%, acute coronary syndrome 21.4%, acute respiratory failure 10%, others 11.4%. Associated diseases: acute renal failure 68.5%, hypertension 60%, diabetes mellitus 37.2%, COPD 10%, infection 91.5%. Using an Utestein-based template, data were collected immediately after each resuscitation, by physicians who have performed CPR. They were previously certified in acute cardiac life support (ACLS).

Results

Seventy cardiopulmonary arrests were recorded in 50 patients. Of these, 49% returned to spontaneous circulation and 4% had hospital discharge. Twelve patients had more than one event. First pulseless rhythm was divided as follows: VF/VT (14.08%), asystole (19.7%), PEA (66.22%). Predictive factors of return to spontaneous circulation in univariate analysis were: time from ICU admission to cardiopulmonary arrest ≤ 7 days (P = 0.03) age <75 years (P = 0.003), time of CPR <18 min (P = 0.0001). In multivariate analysis, only the time from admission to the ICU to cardiopulmonary arrest ≤ 7 days was predictive of a return to spontaneous circulation (P = 0.015, OR 1.19, CI 95% 0.6–5.9).

Conclusion

Survival after CPR in cardiac patients is poor. Considering our population, it is lower than that observed in general hospital patients. These data could help physicians in attempting resuscitation, patients and families in making end-of-life decisions.