Objective: To compare the effectiveness of active compression-decompression (ACD) cardiopulmonary resuscitation (CPR) with a handheld suction device vs standard manual CPR in victims of out-of-hospital cardiopulmonary arrest.
Design: Prospective randomized clinical trial with crossover group design.
Setting: Emergency medical services (EMS) of a large (San Francisco) and medium-sized (Fresno) city in California.
Patients: All normothermic adult victims of out-of-hospital, nontraumatic cardiac arrest on whom CPR was performed by first responders.
Intervention: Patients were randomized to receive either standard manual CPR according to American Heart Association guidelines or ACD CPR, on first-responder contact.
Main outcome measures: Return of spontaneous circulation, admission to the intensive care unit, survival to hospital discharge, and neurological function at hospital discharge.
Results: The ACD group (n = 117 in Fresno; n = 297 in San Francisco) and standard group (n = 136 in Fresno; n = 310 in San Francisco) were similar with regard to demographic and prognostic variables, such as age, witnessed arrest and bystander CPR frequency, and initial cardiac rhythm. Average interval from 911 call activation to EMS responder arrival was 6.4 minutes in Fresno and 4.0 minutes in San Francisco. In Fresno, there was no difference between the ACD group and standard CPR group in return of spontaneous circulation (17% vs 20%; P = .68), hospital admission (16% vs 20%; P = .56), hospital discharge (5% vs 7%; P = .64), or cerebral performance category score at discharge (1.5 vs 1.6; P = .90). Similarly, in San Francisco there was no difference between the ACD group and standard CPR group in return of spontaneous circulation (19% vs 21%; P = .65), hospital admission (13.5% vs 14.5%; P = .79), hospital discharge (4.7% vs 5.5%; P = .80), or cerebral performance category score at discharge (2.2 vs 2.6; P = .31). There was no increase in significant complications associated with the use of ACD CPR.
Conclusion: There was no improvement in outcome with ACD CPR in out-of-hospital cardiac arrest in these two cities. Differences in study design, demographics, EMS systems, response intervals, training, and technique performance may contribute to the lack of improvement in initial resuscitation with ACD CPR compared with previous studies. Future research needs to control these variables to determine the reason for these differences in outcome.