PURPOSE: In spite of medical advances, outcomes remain poor after CPA. We describe predictors that influence mortality in CPA.
METHODS: We conducted a retrospective review of all patients who received CPA in a 669-bed academic institution between July 2000 and October 2005.
RESULTS: Among 728 patients, 476 were male (67.1%), average age was 58 ± 16 years, 349 (51.8%) were Caucasian, 290 (43%) were African American, and 25 (3.7%) were Asian. Admission services were: surgical (42.8%), non-cardiac medical (36.2%) and cardiac medical (16.7%). Settings were: ICU (62.1%), non-monitored floor (19.8%) and monitored non-ICU (16.8%). 50.8% of the CPAs were cardiac. Most were witnessed (91.4%) and occurred on a monitor (84.2%). Average arrest duration was 24 ± 22 minutes. Outcomes: 383 (53.9%) had successful return of spontaneous circulation, 150 (20.5%) survived to hospital discharge. The following were not significant survival predictors: race, sex, age, presence of a monitor or a witness. Predictors of mortality were: arrest etiology (respiratory 65% vs. cardiac 83%; p<0.0001), duration (lower quartile 69% vs. upper quartile 89%; p=0.0019), location (ICU 86% vs. monitored or unmonitored floor 65%; p<0.0001) and admission service (surgical 71% vs. medical or cardiac 85%; p<0.0002).
CONCLUSION: CPA carries a high mortality that is not lessened by monitored settings or by having a witness to the arrest. CPA of respiratory origin carries a lower mortality.
CLINICAL IMPLICATIONS: Information regarding predictors of survival for patients experiencing in-hospital arrest is important to avoid futile cardiopulmonary resuscitation efforts. This information should be incorporated in discussions with families and patients regarding advanced directives.
DISCLOSURE: Soleyah Groves, No Financial Disclosure Information; No Product/Research Disclosure Information