An outcomes analysis of in-hospital cardiopulmonary resuscitation: the futility rationale for do not resuscitate orders.

PURPOSE: Cardiopulmonary resuscitation (CPR) is a frequently performed medical intervention in hospitalized patients who die. Despite the widespread use of do-not-resuscitate (DNR) orders during the last decade, the outcome following CPR appears not to have improved. The key to an improved outcome may be better patient selection. The objective of this study was to determine the hospital survival rate following CPR in the era of DNR orders, and to identify risk factors predictive of hospital survival at a university-affiliated teaching hospital.
MATERIALS AND METHODS: We retrospectively reviewed the code sheets and patient charts of all patients who underwent CPR during a 4-year period from January 1991 to January 1995. Three-hundred-and-eight patients were identified.
RESULTS: CPR was successful in 99 (32%) patients, with 41 (13%) patients surviving to hospital discharge. All the patients who survived were otherwise "healthy" with reversible conditions, who experienced a sudden and unexpected arrhythmic event. No pre-arrest risk factors could clearly distinguish the hospital survivors from the nonsurvivors. The length of the code was 9.4 +/- 4 minutes in the hospital survivors compared with 26.6 +/- 19.1 minutes in the nonsurvivors. Patients whose initial rhythm was either ventricular tachycardia or fibrillation had a better survival rate than patients with other rhythms.
CONCLUSION: DNR protocols do not prevent CPR being performed on patients who are unlikely to survive to hospital discharge. CPR should only be offered to patients who are likely to derive benefit from this intervention.